CN108429776A

CN108429776A - Method for pushing, device, client, interactive device and the system of network object

Info

Publication number: CN108429776A
Application number: CN201710076975.6A
Authority: CN
Inventors: 靳玉康; 马艳玲; 许涵斌
Original assignee: Alibaba Group Holding Ltd
Current assignee: Zhejiang Tmall Technology Co Ltd
Priority date: 2017-02-13
Filing date: 2017-02-13
Publication date: 2018-08-21
Anticipated expiration: 2037-02-13
Also published as: CN108429776B

Abstract

This application discloses a kind of method for pushing of network object and devices, wherein the method includes：Feature buyer is screened based on historical trading, search has associated at least one target signature buyer with the active user of immediate communication platform, and obtain at least one target network object that target signature buyer is accessed, by instant communication client target network object is pushed to active user, since the buyer that can filter out more reliable pushes, the factor of merit of the reliability in push source and the network object of push can be improved, while also improving the safety of transaction；In addition, since the potential incidence relation being utilized between buyer and buyer carries out the push of network object, compared to the push of onrelevant, buyer in application scheme with incidence relation accesses the possibility higher of consolidated network object with active user, the flowing of access that network object after push is increased with this, improves the conversion ratio of the accuracy rate of push and the network object of push.

Description

Method for pushing, device, client, interactive device and the system of network object

Technical field

This application involves technical field of information recommendation, and in particular to the method for pushing of network object, the push of network object Device, a kind of instant communication client, a kind of information interaction device and a kind of supplying system of network object.

Background technology

With the development of internet, the mode of the network push of commodity is more and more various, and the channel of push is also more and more Sample, the shopping record that existing push mode is typically based on user are recommended, and the channel of recommendation is also all traditional canal Road, such as social network sites, forum, blog, short message etc..

A kind of existing commodity push mode is require a high expenditure of energy foundation and target early period of commodity push Relationship between user, usually push only can be directed to good friend, the people of concern or platform user group, cause towards Audient is fixed, and single, the commodity pushed for audient also are difficult to specific aim.

Another commodity push mode is when buyer or seller carry out commodity push, to lack effectively to buyer or seller The mechanism screened is unable to ensure the reliability in push source so that the commodity of push are not inconsistent with user demand, then cause The relatively low problem of the accuracy and conversion ratio of push.

Invention content

In view of the above problems, it is proposed that the application overcoming the above problem in order to provide one kind or solves at least partly State the method for pushing and the pusher of corresponding network object and the supplying system of network object of the network object of problem.

According to the one side of the application, a kind of method for pushing of network object is provided, including：

Historical trading based on transaction platform screens feature buyer, and the transaction platform is used to provide buyer and seller is directed to The transactional services of network object；

Search has associated at least one target signature buyer with the active user of immediate communication platform, and obtains target At least one target network object that feature buyer is accessed, the immediate communication platform support be based on internet between user When exchange message；

By instant communication client the target network object is pushed to active user.

Optionally, there is the active user of the lookup and immediate communication platform associated at least one feature buyer to wrap It includes：

It searches and is at least one feature buyer in set distance range with the active user of immediate communication platform.

Optionally, the lookup and the active user of immediate communication platform are at least one spy in set distance range Levying buyer includes：

Geographic distance between the two is calculated according to the active user and the respective geographical location information of feature buyer；

Extraction geographic distance meets at least one feature buyer of set distance range.

Optionally, it is described the target network object is pushed to active user by instant communication client before, institute The method of stating further includes：

It is sequentially arranged according to the distance of character pair buyer and active user according to by the target network object Sequence.

Optionally, the method further includes：

The request that network object is pushed to active user is received, the request carries the geographical location letter of the active user Breath, the request are generated when active user opens instant communication client or accesses push function entrance.

Optionally, further include：

The geographical location information that feature buyer periodically submits is received, or the transaction address information of the feature buyer is converted For geographical location information.

Optionally, the immediate communication platform shares same system of account with transaction platform；

It is described by instant communication client to active user push the target network object when, the method is also wrapped It includes：

The target seller of the target network object is pushed to the active user.

Optionally, at least one target network object that the acquisition target signature buyer is accessed, the method Further include：

Obtain the target seller that the target network object is provided.

Optionally, further include：

The communication message sent to the target seller by instant communication client is received, and the communication message is turned It is sent to the instant communication client of target signature buyer.

Optionally, described to include to active user's push target network object by instant communication client：

The push page for target network object, the push page are sent to the instant communication client of active user Include transaction entrance, the transaction entrance is operated to the purchase for coping with target network object or operation on probation.

Optionally, the transaction entrance carries buyer's mark of target signature buyer；

The method further includes：

Receive the transaction request sent by accessing the transaction entrance of instant communication client；

Buyer's mark that transaction request carries is extracted, the avail data of this corresponding transaction request is increased into buyer's mark The account data of the target signature buyer of label.

Optionally, further include：

Updated account data is pushed into the corresponding instant communication client of target signature buyer.

Optionally, the historical trading screening feature buyer based on transaction platform includes：

From character network object in the corresponding all buyers of historical trading, according at least one buyer's attribute selection feature Buyer；

Before the historical trading screening feature buyer based on transaction platform, the method further includes：

Character network object is screened according at least one object properties of the network object of historical trading.

Optionally, described to include according at least one buyer's attribute selection feature buyer：

The evaluation score value of each buyer is calculated according at least one buyer's attribute and corresponding attribute weight；

Screening and assessment score value meets the buyer of set interval as feature buyer.

Optionally, at least one object properties of the network object according to historical trading screen character network object packet It includes：

Extraction section network object obtains the object classification label of network of samples object, institute as network of samples object It includes character network object and non-character network object to state object classification label；

The object classification label and object properties of statistical sample network object generate and predict object classification according to object properties The prediction data set of label；

Using the object classification label of each network object of prediction data set prediction history transaction, and according to object Tag along sort extracts character network object.

Optionally, character network object is screened at least one object properties of the network object according to historical trading Before, the method further includes：

The transaction feature data of each network object are extracted from various dimensions, the transaction feature data include trading activity data At least one of with mutual-action behavior data；

Judge the numerical intervals of the transaction feature attribution data, and searches the object category that the numerical intervals correspond to instruction Property, the numerical intervals divide to obtain according to the transaction feature data of all-network object.

From feature seller in the corresponding all-network object of historical trading, feature is screened according at least one object properties Network object；

The network object according to historical trading at least one object properties screening character network object it Before, the method further includes：

From the corresponding all sellers of historical trading, according at least one seller's attribute selection feature seller.

Optionally, described to include according at least one seller's attribute selection feature seller：

Extraction section seller obtains seller's tag along sort of sample seller as sample seller, seller's contingency table Label include feature seller and non-feature sellers；

Seller's tag along sort of statistical sample seller and sell it is attribute, generate according to seller's attribute forecast seller's tag along sort Prediction data set；

Seller's tag along sort of each seller is predicted using the prediction data set, and spy is extracted according to seller's tag along sort Levy seller.

Screening sells the attribute seller for meeting setting seller's condition as feature seller.

According to the another aspect of the application, a kind of method for pushing of network object is provided, including：

Instant communication client is opened according to active user or network object is generated to the access operation of push function entrance Push request；

The push request is sent to server-side；

The target network object of server-side push is received, the target network object is to have associated mesh with active user The network object that mark feature buyer is accessed, historical trading screening of the feature buyer based on transaction platform obtain.

According to the another aspect of the application, a kind of information interacting method is provided, including：

Show push function entrance；

Receive the access operation to push function entrance；

The push request of network object is generated, and is sent to server-side；

The target network object of server-side push is received, the target network object is to have associated mesh with active user The network object that mark feature buyer is accessed, historical trading screening of the feature buyer based on transaction platform obtain；

It shows the target network object and the target seller of the target network object is provided.

Feature buyer is screened based on historical trading；

It searches and is at least one target signature buyer in set distance range with active user, and obtain target signature At least one target network object that buyer is accessed；

The target network object is pushed to active user.

According to the another aspect of the application, a kind of pusher of network object is provided, including：

Feature seller's screening module screens feature buyer for the historical trading based on transaction platform；

Target network object acquisition module has associated at least one for searching with the active user of immediate communication platform A target signature buyer, and obtain at least one target network object that target signature buyer is accessed；

Target network Object Push module, for pushing the target network to active user by instant communication client Object.

Optionally, the target network object acquisition module includes：

Feature buyer searches submodule, is in set distance range with the active user of immediate communication platform for searching At least one feature buyer.

Optionally, the feature buyer lookup submodule includes：

Apart from computation subunit, for calculating two according to the active user and the respective geographical location information of feature buyer Geographic distance between person；

Feature buyer extracts subelement, and at least one feature that set distance range is met for extracting geographic distance is bought Family.

Optionally, described device further includes：

Network object sorting module, for pushing the target network to active user by instant communication client described Before network object, carried out sequentially according to the distance of character pair buyer and active user according to by the target network object Sequence.

Optionally, described device further includes：

Request receiving module is worked as receiving the request for pushing network object to active user described in the request carrying The geographical location information of preceding user, the request is when active user opens instant communication client or accesses push function entrance It generates.

Optionally, further include：

Location information receiving module, the geographical location information periodically submitted for receiving feature buyer, or by the feature The transaction address information of buyer is converted to geographical location information.

Optionally, the immediate communication platform shares same system of account with transaction platform；Described device further includes：

Target seller's pushing module, for pushing the target network to active user by instant communication client described When network object, the target seller of the target network object is pushed to the active user.

Optionally, described device further includes：

Target seller's acquisition module, at least one target network pair for being accessed in the acquisition target signature buyer As when, obtain the target seller that the target network object is provided.

Optionally, further include：

Communication message receiving module disappears for receiving the communication sent to the target seller by instant communication client It ceases, and the communication message is forwarded to the instant communication client of target signature buyer.

Optionally, the target network Object Push module includes：

Page sending submodule is pushed, target network object is directed to for being sent to the instant communication client of active user The push page, the push page include transaction entrance, it is described transaction entrance to cope with target network object purchase grasp Make or try out operation.

Optionally, the transaction entrance carries buyer's mark of target signature buyer；Described device further includes：

Transaction request receiving module is asked for receiving the transaction sent by the transaction entrance for accessing instant communication client It asks；

Buyer's marker extraction module, buyer's mark for extracting transaction request carrying, by this corresponding transaction request Avail data increases to the account data that buyer identifies the target signature buyer of label.

Optionally, further include：

Account data pushing module, for updated account data to be pushed to the corresponding Instant Messenger of target signature buyer Interrogate client.

Optionally, the feature seller screening module includes：

Feature buyer screens submodule, is used for from character network object in the corresponding all buyers of historical trading, according to At least one buyer's attribute selection feature buyer；

Described device further includes：

Character network object screening module, for the historical trading screening feature buyer based on transaction platform it Before, screen character network object according at least one object properties of the network object of historical trading.

Optionally, the feature buyer screening submodule includes：

Score value computation subunit is evaluated, for respectively being bought according at least one buyer's attribute and the calculating of corresponding attribute weight The evaluation score value of family；

Feature buyer screens subelement, and screening and assessment score value meets the buyer of set interval as feature buyer.

Optionally, the character network object screening module includes：

Label acquisition submodule for extraction section network object as network of samples object, and obtains network of samples pair The object classification label of elephant, the object classification label include character network object and non-character network object；

First data acquisition system generates submodule, is used for the object classification label and object properties of statistical sample network object, Generate the prediction data set that object classification label is predicted according to object properties；

Character network object extraction submodule, for each network pair using prediction data set prediction history transaction The object classification label of elephant, and according to object classification tag extraction character network object.

Optionally, described device further includes：

Data extraction module, it is special for being screened at least one object properties of the network object according to historical trading Before levying network object, the transaction feature data of each network object are extracted from various dimensions, the transaction feature data include transaction At least one of behavioral data and mutual-action behavior data；

Numerical intervals determination module, the numerical intervals for judging the transaction feature attribution data, and search the number It is worth the object properties that section corresponds to instruction, the numerical intervals divide to obtain according to the transaction feature data of all-network object.

Optionally, the character network object screening module includes：

Object screens submodule, is used for from feature seller in the corresponding all-network object of historical trading, according at least A kind of object properties screening character network object；

Described device further includes：

Feature seller's screening module, at least one object category in the network object according to historical trading Property screening character network object before, from the corresponding all sellers of historical trading, according at least one seller's attribute selection spy Levy seller.

Optionally, the feature seller screening module includes：

Seller's tag along sort acquisition submodule for extraction section seller as sample seller, and obtains sample seller's Seller's tag along sort, seller's tag along sort include feature seller and non-feature sellers；

The second data set generate submodule, for statistical sample seller seller's tag along sort and sell it is attribute, generate According to the prediction data set of seller's attribute forecast seller's tag along sort；

Feature seller's extracting sub-module, seller's tag along sort for predicting each seller using the prediction data set, And feature seller is extracted according to seller's tag along sort.

Optionally, the feature seller screening module includes：

Seller screens submodule, and the attribute seller for meeting setting seller's condition is sold as feature seller for screening.

According to the another aspect of the application, a kind of instant communication client is provided, including：

Push requests to generate module, for opening instant communication client according to active user or to push function entrance Access operation generates the push request of network object；

Request sending module, for sending the push request to server-side；

Target network object receiving module, the target network object for receiving server-side push, the target network pair As to have the network object that associated target signature buyer is accessed with active user, the feature buyer is based on transaction platform Historical trading screening obtain.

According to the another aspect of the application, a kind of information interaction device is provided, including：

Displaying control, push function entrance；

Wherein, the displaying control is for showing functional entrance；

The displaying functional entrance is used to receive the access operation of active user；

The push that described information interactive device is used to generate network object according to access operation is asked, and is sent to service End；The target network object of server-side push is received, the target network object is to have associated target special with active user The network object that sign buyer is accessed, historical trading screening of the feature buyer based on transaction platform obtain；

The displaying control is additionally operable to show the target network object and provides the target of the target network object Seller.

Buyer's module is screened, for screening feature buyer based on historical trading；

Object acquisition module is bought for searching at least one target signature being in set distance range with active user Family, and obtain at least one target network object that target signature buyer is accessed；

Object Push module, for pushing the target network object to active user.

According to the another aspect of the application, a kind of supplying system of network object is provided, including：

Transaction platform, immediate communication platform, instant communication client；

The immediate communication platform screens feature buyer for the historical trading based on transaction platform；And it searches and is When communication platform active user there is associated at least one target signature buyer, and obtain what target signature buyer was accessed At least one target network object；

The instant communication client is used to receive the target network object that the immediate communication platform is sent and is opened up Show.

According to the embodiment of the present application, feature buyer is screened by being based on historical trading, lookup is worked as with immediate communication platform Preceding user has associated at least one target signature buyer, and obtains at least one target network that target signature buyer is accessed Network object pushes target network object so that can be according to historical trading to buying by instant communication client to active user Family is screened, and after filtering out feature buyer, the target signature buyer relevant with active user is then filtered out, by mesh The target network Object Push of feature buyer's historical trading is marked to active user.

On the one hand, pushed since more reliable buyer can be filtered out, can improve push source reliability with And the factor of merit of the network object of push, while also improving the safety of transaction；On the other hand, since buyer is utilized and buys Potential incidence relation between family carries out the push of network object, compared to the push of onrelevant in background technology, the application side Buyer in case with incidence relation accesses the possibility higher of consolidated network object with active user, after increasing push with this The flowing of access of network object improves the conversion ratio of the accuracy rate of push and the network object of push.

Further, according to different object properties, it can filter out that sales volume bigger, favorable comment be more, commercial quality more has guarantor Hinder, is more favourable, more recommendable network object, then further screening is bought in the buyer of the network object filtered out Family improves the factor of merit of the reliability in push source and the network object of push.

Further, it sells attribute according to different, the seller that favorable comment number is more, cargo exchange hand is high can be filtered out, then Network object is further screened in the network object of the seller filtered out, improve indirectly push source reliability and The factor of merit of the network object of push.

Further, it due to providing more convenient and more high-conversion rate push mode for buyer, improves buyer and participates in pushing away The liveness sent increases the line duration of buyer and online amount, increases the communication between buyer, while increasing for seller The channel of push.

Above description is only the general introduction of technical scheme, in order to better understand the technological means of the application, And can be implemented in accordance with the contents of the specification, and in order to allow above and other objects, features and advantages of the application can It is clearer and more comprehensible, below the special specific implementation mode for lifting the application.

Description of the drawings

By reading the detailed description of hereafter preferred embodiment, various other advantages and benefit are common for this field Technical staff will become clear.Attached drawing only for the purpose of illustrating preferred embodiments, and is not considered as to the application Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings：

Fig. 1 shows a kind of schematic diagram of the push scene of network object according to the application；

Fig. 2 shows the flow charts according to the method for pushing embodiment of network object of the embodiment of the present application one a kind of；

Fig. 3 shows a kind of flow chart of the method for pushing embodiment of network object according to the embodiment of the present application two；

Fig. 4 shows a kind of flow chart of the method for pushing embodiment of network object according to the embodiment of the present application three；

Fig. 5 shows a kind of flow chart of information interacting method embodiment according to the embodiment of the present application four；

Fig. 6 shows a kind of flow chart of the method for pushing embodiment of network object according to the embodiment of the present application five；

Fig. 7 shows the push process schematic of network object in the example of the embodiment of the present application；

Fig. 8 shows the push process schematic of network object in the example of the embodiment of the present application；

Fig. 9 shows a kind of structure diagram of the pusher of network object according to the embodiment of the present application six；

Figure 10 shows a kind of structure diagram of instant communication client according to the embodiment of the present application seven；

Figure 11 shows a kind of structure diagram of the pusher of network object according to the embodiment of the present application nine；

Figure 12 shows a kind of structure diagram of the supplying system of network object according to the embodiment of the present application ten.

Specific implementation mode

The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although showing the disclosure in attached drawing Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure Completely it is communicated to those skilled in the art.

The application can apply in the Push Service of instant message applications, social network sites, social software etc., for example, root According to the transaction data on electric business platform, filter out that liveness is high, evaluation frequency is high or the buyers of other features, find with currently User has the associated buyer filtered out, then the commodity bought with associated buyer is pushed to active user.

In the application, transaction platform can be the platform concluded the transaction by network, be directed to for providing buyer and seller The transactional services of network object, the network object of transaction are the arbitrary object that can be accessed by network, and network object can be with It is at least one of commodity, picture, video, music, novel etc..

Buyer is the user that network object is bought in trading activity, and seller is the use that network object is sold in trading activity Family.

Historical trading is the transaction event completed in the history recorded on transaction platform, and historical trading is on transaction platform The various information of each secondary transaction event can be specifically recorded, such as may include mark, the transaction of the seller and buyer of transaction Network object, the time of transaction, seller and various user informations of buyer etc..Feature buyer is to have specific spy in historical trading The buyer of sign, such as buyer that liveness is high, comment on that the buyer more than number, sales volume is high and evaluates liveness in the buyer of high commodity The high-quality buyer such as high buyer.

Immediate communication platform is supported to be based on interconnection between user to provide the communication platform of exchange message immediately based on network Net exchanges message immediately, and the immediate communication platform that can be integrated on transaction platform can also be not straight with transaction platform Connect the immediate communication platform of relationship.Instant communication client is a part for immediate communication platform, between instant communication client Message exchange is completed by immediate communication platform.

Referring to Fig.1, a kind of schematic diagram of the push scene of network object according to the application is shown.Specifically pushed Cheng Wei, immediate communication platform obtain the corresponding data of historical trading of transaction platform, according to historical trading, filter out feature and buy Family, searched in all feature buyers has associated at least one target signature buyer with active user, obtains target signature At least one target network object that buyer is accessed, by least one target network object by instant communication client to working as Preceding user.For example, according to the purchase data of transaction platform, at least one high-quality buyer is filtered out, the position of active user is obtained With the position of high-quality buyer, filter out with high-quality buyer of active user's distance in setting range as the high-quality buyer of target, The history purchase data for obtaining the high-quality buyer of target obtains at least commodity that the high-quality buyer of target bought, will an at least quotient Product are pushed to active user.

With reference to Fig. 2, a kind of flow of the method for pushing embodiment of network object according to the embodiment of the present application one is shown Figure, this method can specifically include following steps：

Step 101, the historical trading based on transaction platform screens feature buyer.

Transaction platform is used to provide buyer and seller is directed to the transactional services of network object, in the historical trading of transaction platform In, record has the buyer and seller of each secondary transaction event, is recorded accordingly according to historical trading and determines that user is in each transaction Seller or buyer.Historical trading screening feature buyer based on transaction platform refer to according to the data of historical trading to buyer into Row screening, filters out the buyer for meeting specified conditions, such as the buyer more than transaction count, transaction duration time short buyer, steps on Buyer etc. more than record buyer often, comment buyer often, good friend's quantity, specifically can also be and meets above multiple The buyer of part, such as transaction count are more than setting number and comment on the buyer that number is more than setting number.The data of historical trading May include network object, the buyer of transaction and the mark of seller of transaction, the time of transaction, each attribute of buyer, seller The data such as each attribute.

Specifically, a kind of mode of screening feature buyer can be belonged to according at least one buyer from all buyers Property screening feature buyer, for example, buyer's attribute may include the registion time of buyer, it is user gradation, transaction count, one nearest Transaction count, line duration, comment number, chat messages item number, good friend's quantity, number of contacts of the moon etc..According at least one A kind of mode of kind buyer attribute selection feature buyer can reach respective given threshold screening according at least one attribute Go out feature buyer, any suitable screening mode specifically may be used, the embodiment of the present application is not restricted this.

Step 102, search has associated at least one target signature buyer with the active user of immediate communication platform, and Obtain at least one target network object that target signature buyer is accessed.

Message is exchanged based on internet immediately between immediate communication platform support user, active user is immediate communication platform User, if immediate communication platform and transaction platform share same system of account, the user of immediate communication platform can be handed over The buyer or seller of easy platform, if two sets of different system of account, then the user may have another pair to answer in transaction platform Account, it is also possible to which there is no corresponding accounts, the only potential user of transaction platform.

It refers to that user and at least one attribute of buyer are related, such as can be that active user has association to feature buyer The distance between current location is in set distance range or user and buyer belong to same user's classification, either User is consistent with the preference of access network object of buyer or the network object of the collection of user and buyer is part or all of It is identical etc..Target signature buyer is to have associated feature buyer with the active user of immediate communication platform.

Being searched in feature buyer with the active user of immediate communication platform there is associated at least one target signature to buy Family, can specifically obtain at least one attribute of active user, according at least one attribute of active user, search all features Relevant at least one attribute in buyer, determines that corresponding feature buyer is target signature buyer.

After finding at least one target signature buyer, at least one target network that target signature buyer is accessed is obtained Network object can specifically include all or part of target network object that target signature buyer accessed, such as may include mesh The commodity etc. of commodity, collection that mark feature buyer bought, can specifically include any suitable network object, and the application is implemented Example is without limitation.

For example, obtaining the current location of the active user of immediate communication platform, the current location setting of active user is searched Feature buyer in range is target signature buyer, and it is target network pair to obtain the network object that target signature buyer is accessed As.

Step 103, the target network object is pushed to active user by instant communication client.

In at least one target network object of acquisition, by part or all of target network object according to random sequence or According to the sequence of certain rule, target network object is pushed to active user by instant communication client, pushes target network When the information of object, can also carry the corresponding target signature buyer of target network object in the mark of immediate communication platform or The shortcut link that can initiate chat, can be specifically any suitable push mode, and the embodiment of the present application is without limitation.

Specifically, a kind of mode of push can send to be directed to target network to the instant communication client of active user The page link of network object can enter the displayed page of target network object according to page link, or according to page link The chat window with the seller user of target network object can be entered.

For example, according to target signature buyer at a distance from active user, according to the sequence of distance from the near to the remote, first push is close At least one target network object of the target signature buyer at place, then push at least one target of the target signature buyer of distant place Network object also shows target network object on the instant communication client of active user according to sequence from the near to the remote.

In a preferred embodiment of the present application, the immediate communication platform shares same system of account with transaction platform； When pushing the target network object to active user by instant communication client described in the execution step 103, may be used also To include：

Sub-step S1 pushes the target seller of the target network object to the active user.

If immediate communication platform is used in conjunction with same set of system of account with transaction platform, user both may be used by current account With log-in instant communication platform, and transaction platform can be logged in.Target seller is the seller for providing target network object, to current When user pushes target network object, the target seller of target network object can also be pushed to active user, it specifically can be with It is to be pushed to target seller in the mark of immediate communication platform, electronic business card or the quick entrance etc. of establishing chat window to work as Target seller can be specifically pushed to active user by any suitable mode by preceding user, and the embodiment of the present application does not do this Limitation.

In a preferred embodiment of the present application, described in the step 102 obtain target signature buyer accessed to When a few target network object, can also include：

Sub-step S2 obtains the target seller for providing the target network object.

When obtaining at least one target network object that target signature buyer is accessed, offer target network can also be obtained The target seller of network object, such as when acquisition target signature buyer purchase at least one commodity, obtain at least one quotient of purchase The respective seller of product is as target seller.

In a preferred embodiment of the present application, can also include：Step 104, receive by instant communication client to The communication message that the target seller sends, and the communication message is forwarded to the instant messaging client of target signature buyer End.

When active user sends communication message by instant communication client to target seller, server receives the logical of transmission Message is interrogated, and communication message is forwarded to the instant communication client of the corresponding target signature buyer of target network object.

With reference to Fig. 3, a kind of flow of the method for pushing embodiment of network object according to the embodiment of the present application two is shown Figure, this method can specifically include following steps：

Step 201, character network object is screened according at least one object properties of the network object of historical trading.

Object properties refer to the characteristic attribute that network object has, and may include the relevant data letter of transaction of network object The attribute information etc. that breath or network object have.Such as object properties may include the number of deals of network object, network object By comment number, network object containing picture by comment number, favorable comment number, difference commenting number, seller reply comment number, sell area Quantity, buyer's number, sells number and accounts for seller and always sell several accountings, can also include to network object various on-line time The evaluation information etc. of dimension.It can specifically include any suitable object properties in the application, the embodiment of the present application does not make this Limitation.The network object filtered out according at least one object properties of the network object of historical trading is characterized network object.

In practical application, according to different object properties, can filter out sales volume bigger, favorable comment be more, commercial quality more Secure, more favourable, more recommendable network object, is then further sieved in the buyer of the network object filtered out Choosing improves the factor of merit of the reliability in push source and the network object of push.

It is set specifically, screening character network object can reach given threshold or meet according at least one object properties Fixed condition is used as screening foundation, such as can be that one or more object properties reach respective given threshold or meet and set Fixed condition, can be specifically any suitable screening foundation, and the embodiment of the present application is without limitation.

For example, a kind of screening mode can be：It filters out comment number and reaches the first given threshold, and favorable comment number reaches second Given threshold, and sell the commodity that regional quantity reaches third given threshold, wherein the first given threshold, the second given threshold Can be any suitable setting value with third given threshold, the embodiment of the present application is without limitation.

For example, another screening mode can be：It obtains the comment number of commodity, favorable comment number, sell quantity, according to comment Number, sells the interval range that quantity respectively belongs at favorable comment number, determines the object properties of commodity, wherein object properties are to comment Number, sells what quantity was divided at favorable comment number, judges whether the object properties of commodity meet sets requirement, filters out symbol The commodity of sets requirement are closed, specific sets requirement can be configured according to actual needs, and the embodiment of the present application does not limit this System.

In a preferred embodiment of the present application, described in the step 201 extremely according to the network object of historical trading A kind of few object properties screen character network object and may include：

Sub-step S3, extraction section network object obtain the object point of network of samples object as network of samples object Class label.

Sub-step S4, the object classification label and object properties of statistical sample network object generate pre- according to object properties Survey the prediction data set of object classification label.

Sub-step S5, using the prediction data set prediction history transaction each network object object classification label, And according to object classification tag extraction character network object.

Object classification label is label for classifying according to different dimensions to network object, in the application, object Tag along sort includes character network object and non-character network object, and character network object is the network object according to historical trading At least one object properties meet the network object of setting condition, other network objects for not meeting setting condition are non-feature Network object.

For example, can be the label that commodity add best buy or inferior goods according to the comment of commodity, for example, specifically may be used Using be favorable comment be more than 9 percent commodity as best buy, it is inferior goods that difference, which comments commodity more than 90 percent,.

The extraction section network object in overall network object obtains network of samples object as network of samples object Object classification label, the object classification label of network of samples object can be that is manually added can also be the rule according to setting It generates.

Predictive data set is combined into the number for the object classification label that can predict other network objects in addition to network of samples object According to set, the prediction data set of object classification label is specially predicted according to object properties.For example, Naive Bayes Classifier It is exactly a kind of prediction data set.In machine learning, Naive Bayes Classifier be it is a series of with assume between feature it is strong solely It makes with the simple probability grader based on Bayes' theorem.The principle of classification of Bayes classifier is by certain object Prior probability calculates its posterior probability using Bayesian formula, i.e. the object belongs to certain a kind of probability, and selection has maximum The class of posterior probability is as the class belonging to the object.Naive Bayesian is a kind of popular method of text classification, and text classification is The problem of file generic or other (such as spam, legitimacy, sport or politics) are judged characterized by word frequency.It is logical Pretreatment appropriate is crossed, it can mutually be competed with this field more advanced method (including support vector machines).

Corresponding object properties under each object classification label are counted respectively, distinguish statistical nature network object pair in the application The object properties and the corresponding object properties of non-character network object answered, according to the distribution feelings of object classification label and object properties Condition generates prediction data set.For example, using Naive Bayes Classifier, sample is determined according to the object properties of network of samples object The object classification label of present networks object calculates probability of occurrence of each object classification label in network of samples object and right Each object properties, which divide, estimates the conditional probability of each object classification label, and result is recorded, and generates for network The Naive Bayes Classifier that object is classified, that is, predict the prediction data set of object classification label.

The object properties for obtaining each network object of historical trading predict other network objects according to prediction data set Object classification label, then according to object classification tag extraction character network object, extracting object tag along sort in the application It is characterized the network object of network object.

For example, extracting sample commodity from all commodity, the tag along sort best buy and poor quality of sample commodity are obtained Commodity count corresponding at least one object properties under label high-quality and inferior, prediction number are generated according to the result of statistics respectively According to set, show that other commodity belong to the probability of best buy and belong to the probability of inferior goods using NB Algorithm, According to belonging to the probability of best buy and belonging to the probability of inferior goods, the factor of merit of other commodity is obtained after weighted calculation, is selected The commodity for taking out the setting number of the highest factor of merit are characterized network object, or select the commodity that the factor of merit reaches sets requirement It is characterized network object.

In a preferred embodiment of the present application, described in the step 201 extremely according to the network object of historical trading Before a kind of few object properties screening character network object, can also include：

Sub-step S6 extracts the transaction feature data of each network object from various dimensions.

Sub-step S7, judges the numerical intervals of the transaction feature attribution data, and searches the numerical intervals correspondence and refer to The object properties shown.

Transaction feature data are used to be directed to feature of the network object in multiple historical trading from the description of some dimension, Can be various evaluation data present in transaction.

Specifically, for the transactional stage that primary transaction is being merchandised from reaching the standard grade completing, to buyer's evaluation network object or Seller replys the interactive stage of comment, produces a variety of data, and a variety of behaviors participated in by buyer or seller can be divided into transaction Behavior and mutual-action behavior, corresponding generation trading activity data and mutual-action behavior data, for example, the transaction row that transactional stage generates Can have for data and sell commodity number, sell commodity number selling proportion in sum, sell area in all commodity Number (such as unit of counties and districts), on-line time, buyer's number etc., mutual-action behavior data can have comment number, containing picture Number of reviews, favorable comment number, seller reply number of reviews etc..Specific data class can be set according to actual scene and needs Fixed, the application is not limited this.

Corresponding object properties, such as favorable comment number can be preset 90 percent to a hundred percent for each numerical intervals The corresponding object properties of numerical intervals be favorable comment commodity, trading volume is in 10,000 or more the corresponding object properties of numerical intervals High sales volume commodity can not limit this according to the specific setting of difference of specific object properties, the application in the concrete realization System.

The transaction feature data that each network object is extracted from various dimensions, judge the numerical intervals of transaction feature attribution data, Wherein, numerical intervals divide to obtain according to the transaction feature data of all-network object.Then numerical intervals are searched and correspond to instruction Object properties.

Sub-step S8, from feature seller in the corresponding all-network object of historical trading, according at least one object category Property screening character network object.

Described in the step 201 character network is screened according at least one object properties of the network object of historical trading Before object, can also include：

Sub-step S9, from the corresponding all sellers of historical trading, according at least one seller's attribute selection feature seller.

Feature seller is to sell the attribute seller for meeting sets requirement or going out according to certain Rules Filtering.It is to sell to sell attribute The standby attribute information of furniture can specifically include seller's grade, historical trading sum, history favorable comment transaction amount, historical review Commodity number, the seller's customer service number etc. that number, history were commented containing picture comment number, seller's commodity number, seller by difference are specific to wrap Include it is any suitable sell attribute, the embodiment of the present application is not restricted this.

It from the corresponding all sellers of historical trading, is sold according at least one attribute, filters out at least one and sell family members Property meets the seller of respective sets requirement as feature seller.Again from feature seller in historical trading corresponding all feature nets In network object, character network object is screened according at least one object properties, according at least one from all character network objects The mode of kind object properties screening character network object is consistent with other embodiment description, does not repeat separately.

For example, according to historical trading data, the seller that favorable comment number is more, cargo exchange hand is high is filtered out, obtains the institute of seller There are commodity, according to the object properties of commodity, filters out the commodity more than favorable comment number and more than comment sum.

In practical application, sells attribute according to different, the seller that favorable comment number is more, cargo exchange hand is high can be filtered out, Then it is further screened in the network object of the seller filtered out, improves reliability and the push in push source indirectly Network object the factor of merit.

In a preferred embodiment of the present application, according at least one seller's attribute selection spy described in the sub-step S7 Levying seller may include：

Sub-step S71, extraction section seller obtain seller's tag along sort of sample seller as sample seller.

Sub-step S72, seller's tag along sort of statistical sample seller and sell it is attribute, generation sold according to seller's attribute forecast The prediction data set of family's tag along sort.

Sub-step S73, seller's tag along sort of each seller is predicted using the prediction data set, and is classified according to seller Tag extraction feature seller.

Seller's tag along sort is label for classifying according to different dimensions to seller, in the application, seller's classification Label includes feature seller and non-feature sellers, and feature seller is to sell attribute symbol according at least one of the seller of historical trading The seller to impose a condition is closed, other sellers for not meeting setting condition are non-feature seller.

For example, can be the label that seller adds high-quality seller or seller inferior according to the history favorable comment sum of seller, than Such as, can be specifically the seller that favorable comment is more than 9 percent it be high-quality seller, it is poor quality that difference, which comments seller more than 90 percent, Seller.

The extraction section seller in whole sellers obtains seller's tag along sort of sample seller, sample as sample seller Seller's tag along sort of seller can be manually add can also be according to setting rule generate.

The predictive data set of prediction seller's tag along sort is combined into the seller point that can predict other sellers in addition to sample seller The data acquisition system of class label, specially according to the prediction data set of seller's attribute forecast seller's tag along sort.For example, support to Amount machine (full name in English：Support Vector Machine, English abbreviation：SVM) it is exactly a kind of prediction data set.In machine In study, support vector machines is to classify and the supervised learning model of analysis data in regression analysis and relevant study calculation Method.One group of training example is given, each training example is marked as belonging to one or the other in two classifications, SVM training Algorithm creates a model by new example allocation to one of two classifications, becomes non-probability binary linearity grader. SVM models are the points being expressed as example in space, and mapping in this way allows for the example of independent classification by as wide as possible apparent It is spaced apart.Then, new example is mapped to the same space, and based on them falls the which side at interval come belonging to predicting Classification.

Count respectively under each seller's tag along sort it is corresponding sell it is attribute, in the application distinguish statistical nature seller it is corresponding Sell attribute and non-feature seller it is corresponding sell it is attribute, according to seller's tag along sort and sell attribute distribution situation generate it is pre- Measured data set.For example, using support vector machines, according to the seller's contingency table for selling attribute determining sample seller of sample seller Label, the frequency of occurrences of seller's tag along sort of statistical sample seller and sell the attribute frequency of occurrences, using support vector machines into Row training obtains the model file classified for seller, that is, the prediction data set of prediction seller's tag along sort.

It is attribute to obtain selling for each seller of historical trading, according to prediction data set, predicts the seller point of other sellers Then class label extracts feature seller according to seller's tag along sort, extraction seller's tag along sort is characterized seller's in the application Seller.

For example, extracting sample seller from all sellers, the high-quality seller of tag along sort and the poor quality of sample seller are obtained Seller, count respectively corresponding at least one under label high-quality and inferior sell it is attribute, according to the result of statistics generate prediction number According to set, prediction data set is obtained using support vector machines, is selected in all sellers according to prediction data set high-quality Seller.

Sub-step S74, screening sell the attribute seller for meeting setting seller's condition as feature seller.

In all sellers, according to sell it is attribute meet setting seller conditional filtering feature seller, specifically can be according to extremely Few one kind sell it is attribute meet respective setting seller conditional filtering feature seller, can specifically set seller according to actual needs Condition, the embodiment of the present application are without limitation.

For example, filtering out the seller that favorable comment number is more, commodity exchange hand is high, specifically given threshold can be reached with favorable comment number, and Commodity exchange hand reaches the seller of given threshold.

Step 202, from character network object in the corresponding all buyers of historical trading, according at least one buyer's attribute Screen feature buyer.

The character network object screened corresponding all buyers in historical trading are obtained, therefrom according at least one buyer Attribute selection feature buyer, the mode for screening feature buyer is consistent with other embodiment description, does not repeat separately.

In a preferred embodiment of the present application, according at least one buyer's attribute selection spy described in the step 202 Levying buyer may include：

Sub-step S8 calculates the evaluation score value of each buyer according at least one buyer's attribute and corresponding attribute weight.

Sub-step S9, screening and assessment score value meet the buyer of set interval as feature buyer.

Whether the evaluation score value of buyer is for evaluating buyer can be as the score value of feature buyer.Each buyer's attribute setup There is corresponding attribute weight, extract the attribute weight of various buyer's attributes, according at least one buyer's attribute and corresponding category Property weight obtains the evaluation score value of each buyer to the corresponding score value weighted calculation of buyer's attribute, for example, by the shopping number of buyer, It is added after online hours, comment number weighting and obtains the evaluation score value of buyer.Then, buyer of the score value in set interval will be evaluated As feature buyer.

Step 203, it searches and is at least one feature in set distance range with the active user of immediate communication platform Buyer, and obtain at least one target network object that target signature buyer is accessed.

The location information that the location information and feature buyer reported according to active user reports, by calculate active user and The distance of feature buyer finds at least one feature being in the active user of immediate communication platform in set distance range Buyer.

Specifically, active user submits current GPS (full name in English to server by instant communication client： Global Positioning System, Chinese name：Global positioning system) data or transaction address conversion geographical location Information, the geographical location information that feature buyer timing is converted to server update GPS data or transaction address, server foundation GPS data or geographical location information calculate separately out the distance between active user and each feature buyer, then select with immediately The active user of communication platform is at least one feature buyer in set distance range, can be specifically any suitable set Set a distance range, the embodiment of the present application are without limitation.

At least one feature buyer searched is target signature buyer, and acquisition target signature buyer is accessed at least one Target network object, the mode for specifically obtaining at least one target network object is consistent with other embodiment, does not repeat separately.

The current use with immediate communication platform is searched in a preferred embodiment of the present application, described in the step 203 At least one feature buyer that family is in set distance range may include：

Sub-step S10 is calculated between the two according to the active user and the respective geographical location information of feature buyer Geographic distance.

Sub-step S11, extraction geographic distance meet at least one feature buyer of set distance range.

Geographical location information can be the information in the geographical location of gps data information or transaction address conversion.It is current to use After family and feature buyer report geographical location information, calculate geographic distance between the two, extraction geographic distance meet setting away from At least one feature buyer from range, set distance range can be set according to actual needs, the embodiment of the present application to this not It is limited.

Can also be that extraction geographical location information belongs to the active user of the same area and feature is bought in practical application Family, for example Haidian is belonged to, can also be that working as in the value range for belonging to setting longitude and latitude is searched according to geographical location information Preceding user and feature buyer, specifically may be used any suitable mode, and the embodiment of the present application is not restricted this.

In a preferred embodiment of the present application, can also include：

Sub-step S12 receives the request that network object is pushed to active user.

Active user sends out the request of request push network object, and request carries the geographical location information of active user, asks Asking can generate when active user opens instant communication client or accesses push function entrance, and server is received to be used to current Family pushes the request of network object, chooses target network object in a manner of according to the push in the application, is pushed to active user.

In a preferred embodiment of the present application, can also include：

Sub-step S13 receives the geographical location information that feature buyer periodically submits, or by the transaction of the feature buyer Location information is converted to geographical location information.

Feature buyer periodically submits geographical location information, or the transaction address information of the feature buyer is converted to geography Location information, can specifically include the GPS data of instant communication client acquisition, and the transaction address information that transaction platform obtains turns The geographical location information changed.

Step 204, according to by the target network object according to the distance of character pair buyer and active user into Row sequentially sorts.

It is corresponding to feature buyer according to distance according to the distance of the active user and each feature buyer that are calculated Target network object is sequentially sorted, specifically can will distance it is close come front, can also will distance it is remote come front, The application is without limitation.

Specifically, when being ranked up to target network object, it is corresponding all target network object acquisitions can be directed to Then feature buyer sequentially sorts to target network object according to distance at a distance from active user, follow-up to carry out It can be pushed according to the sequence of sequence when the push of target network object, or be opened up according to the sequence of sequence after push Show.

Step 205, the push page for target network object is sent to the instant communication client of active user.

It is the page for pushing target network object to push the page, generates the push page for target network object, pushes away It includes transaction entrance to send the page, by that can carry out the operation such as buying or try out to network object to transaction entry operation, this In application, transaction entrance is operated to the purchase for coping with target network object or operation on probation.After generating the push page, used to current The instant communication client at family sends the push page generated, for the instant communication client displaying push page of active user Face.

In a preferred embodiment of the present application, the transaction entrance carries buyer's mark of target signature buyer；Also May include：

Sub-step S14 receives the transaction request sent by accessing the transaction entrance of instant communication client.

Sub-step S15, buyer's mark that extraction transaction request carries, the avail data of this corresponding transaction request is increased The account data of the target signature buyer of label is identified to buyer.

Buyer is identified as the mark of the corresponding target signature buyer of target network object, that is, pushes mesh to active user Mark the buyer of network object.Avail data refers to needing to distribute to target signature buyer in the income that seller obtains in this transaction Some, such as in commodity transaction seller obtain income a part.The account data of target signature buyer is target The account balance of feature buyer.

By accessing the transaction entrance of instant communication client, transaction request can be sent out to target network object, is received Transaction request, buyer's mark that extraction transaction request carries, extracts the avail data of the generation of this transaction request, increases to and buy In the account data of the target signature buyer of family's mark label.For example, carrying target signature buyer in the transaction request of commodity Mark, the part of total revenue that extraction commodity generate after striking a bargain, and by income increase to buyer identify label target it is special In the account balance for levying buyer.

Further, the liveness that buyer participates in push is improved, the line duration of buyer and online amount is increased, increases Communication between buyer, while being the channel that seller increases push.

In a preferred embodiment of the present application, can also include：

Updated account data is pushed to the corresponding instant communication client of target signature buyer by sub-step S16.

After account data update, account data is pushed to the instant communication client of corresponding target signature buyer, with notice Target signature buyer's account data changes.

According to the embodiment of the present application, by screening character network object according to different object properties, from character network pair In all buyers of elephant, according at least one buyer's attribute selection feature buyer, the active user with immediate communication platform is searched At least one feature buyer in set distance range, and obtain at least one target network that target signature buyer is accessed Network object is sequentially sorted according to by target network object according to the distance of character pair buyer and active user, to The instant communication client of active user sends the push page for target network object so that can be according to historical trading pair Network object is screened, and after filtering out character network object, feature buyer is screened in the buyer of character network object, according to Geographic distance between active user and feature buyer further filters out and pushes at least one of network object for active user Apart from close target signature buyer, by least one target network Object Push of target signature buyer to active user.

On the one hand, since active user and push source distance are closer, the network object that nearby users are recommended is more likely It is that active user is required, not only increases the factor of merit of the network object of push, and the buyer for participating in push can profit With the commodity for apart from advantage being on the spot active user's displaying push；On the other hand, latent between buyer and buyer due to being utilized The push of network object is carried out in incidence relation, compared to the push of onrelevant in background technology, has in application scheme and closes The buyer of connection relationship accesses the possibility higher of consolidated network object with active user, and network object after pushing is increased with this Flowing of access improves the conversion ratio of the accuracy rate of push and the network object of push.

With reference to Fig. 4, a kind of flow of the method for pushing embodiment of network object according to the embodiment of the present application three is shown Figure, this method are applied in instant communication client, and following steps are can specifically include：

Step 301, instant communication client is opened according to active user or the access operation of push function entrance is generated The push of network object is asked.

The method for pushing of the application can depending on the user's operation trigger under several scenes, such as active user opens and is When telecommunication customer end when, generate to active user push network object request, with it is further obtain push network object；Or It is that push function entrance is shown on instant communication client, it is raw when receiving access operation of the user to push function entrance At the request for pushing network object to active user.It can also be triggered under other scenes, the application is not limited this.

Step 302, the push request is sent to server-side.

Push request is sent to server, so that network object is pushed to by server according to push request sends out push The instant communication client of request.

Step 303, the target network object of server-side push is received.

Target network object is the network object accessed with associated target signature buyer with active user, the spy It levies historical trading screening of the buyer based on transaction platform to obtain, concrete mode is consistent with other embodiment description, does not repeat separately.Clothes Target network Object Push to the instant communication client of active user, is received the target network object of push by business end.

According to the embodiment of the present application, by opening instant communication client according to active user or to push function entrance Access operation generates the push request of network object, sends the push request to server-side, receives the target of server-side push Network object, target network object are described to have the network object that associated target signature buyer is accessed with active user Historical trading screening of the feature buyer based on transaction platform obtains so that can be screened, is sieved to buyer according to historical trading After selecting feature buyer, the target signature buyer relevant with active user is then filtered out, target signature buyer is gone through The target network Object Push of history transaction is to active user.

With reference to Fig. 5, a kind of flow chart of information interacting method embodiment according to the embodiment of the present application four, the party are shown Method can specifically include following steps：

Step 401, push function entrance is shown.

Push function entrance is shown on instant communication client, is accessed operation to push function entrance, specifically may be used By with showing that push function entrance, the embodiment of the present application are without limitation in the form of any suitable.

Step 402, the access operation to push function entrance is received.

The access operation to push function entrance is monitored, the access operation to push function entrance is received.

Step 403, the push request of network object is generated, and is sent to server-side.

The request for generating push network object, sends the requests to server-side.

Step 404, the target network object of server-side push is received.

Target network object is with active user there is the network object that is accessed of associated target signature buyer, feature to buy Historical trading screening of the family based on transaction platform obtains, and concrete mode is consistent with other embodiment description, does not repeat separately.Server-side By target network Object Push to the instant communication client of active user, the target network object of push is received.

Step 405, it shows the target network object and the target seller of the target network object is provided.

Target network object and the target seller for providing target network object are shown, can be specifically instant The push page and the target seller of target network object, Huo Zhezhi are shown in the chat window of telecommunication customer end or other windows The chat window of displaying and the target seller of target network object is connect, and shows target network object wherein, is specifically shown Mode can be any suitable mode, and the embodiment of the present application is without limitation.

According to the embodiment of the present application the access operation to push function entrance, life are received by showing push function entrance It is asked at the push of network object, and is sent to server-side, receive the target network object of server-side push, target network object For there is the network object that associated target signature buyer is accessed, history of the feature buyer based on transaction platform with active user Transaction screening obtains, and shows the target network object and provides the target seller of the target network object so that can be with Buyer is screened according to historical trading, after filtering out feature buyer, is then filtered out relevant with active user Target signature buyer, by the target network Object Push of target signature buyer's historical trading to active user.

With reference to Fig. 6, a kind of flow of the method for pushing embodiment of network object according to the embodiment of the present application five is shown Figure, can specifically include following steps：

Step 501, it is based on historical trading and screens feature buyer.

In historical trading, record has the buyer and seller of each secondary transaction event, is recorded accordingly according to historical trading really User is seller or buyer in fixed each transaction.Refer to the data according to historical trading based on historical trading screening feature buyer Buyer is screened, filters out the buyer for meeting specified conditions, such as the buyer more than transaction count, transaction duration time are short Buyer etc. more than buyer more than buyer, login times, comment buyer often, good friend's quantity, specifically can also be meet with The buyer of upper multiple conditions, such as transaction count are more than setting number and comment on the buyer that number is more than setting number.History is handed over Easy data may include network object, the buyer of transaction and the mark of seller of transaction, the time of transaction, each kind of buyer Property, the data such as each attribute of seller.

Step 502, it searches and is at least one target signature buyer in set distance range with active user, and obtain At least one target network object that target signature buyer is accessed.

The location information that the location information and feature buyer reported according to active user reports, by calculate active user and The distance of feature buyer finds at least one feature buyer being in active user in set distance range.

Specifically, active user submits current GPS (full name in English to server：Global Positioning System, Chinese name：Global positioning system) data or transaction address conversion geographical location information, feature buyer timing to Server update GPS data or the geographical location information of transaction address conversion, server are believed according to GPS data or geographical location Breath calculates separately out the distance between active user and each feature buyer, then selects and be in set distance range with active user Interior at least one feature buyer, can be specifically any suitable set distance range, and the embodiment of the present application does not limit this System.

Step 503, the target network object is pushed to active user.

In at least one target network object of acquisition, by part or all of target network object according to random sequence or According to the sequence of certain rule, target network object is pushed to active user, it, can be with when pushing the information of target network object It carries the mark of the corresponding target signature buyer of target network object or the shortcut link of chat can be initiated, can be specifically Any suitable push mode, the embodiment of the present application are without limitation.

Specifically, a kind of mode of push can be the page chain sent to active user for target network object It connects, the displayed page of target network object can be entered according to page link, or can be entered according to page link and target The chat window of the seller user of network object.

For example, according to target signature buyer at a distance from active user, according to the sequence of distance from the near to the remote, first push is close At least one target network object of the target signature buyer at place, then push at least one target of the target signature buyer of distant place Network object, and show target network object according to sequence from the near to the remote.

According to the embodiment of the present application, feature buyer is screened by being based on historical trading, searches and is in setting with active user At least one target signature buyer in distance range, and obtain at least one target network pair that target signature buyer is accessed As pushing target network object to active user so that can be screened to buyer according to historical trading, filter out feature and buy After family, the target signature buyer closely located with active user is then filtered out, by the target of target signature buyer's historical trading Network object is pushed to active user.

To make those skilled in the art more fully understand the application, one kind of the application is tested below by way of specific example Card method illustrates.

Referring to Fig. 7, the push process schematic of network object in an example of the embodiment of the present application is shown.

Step 1, high-quality seller's screening is carried out；

There are many features by high-quality seller, such as agree to comment more, cargo exchange hand height of number etc..We utilize high-quality seller's Feature classifies to seller, more high-quality seller in screening.According to the transaction characteristics of seller, most characteristic N number of goods is chosen Object, and obtain nearest shopping buyer's list of N number of cargo.

It can specifically include：

Sub-step 1, the characteristic of selected section seller are complained according to seller's history and are recorded, handmarking seller's type, Such as：High-quality seller, middle matter seller, seller inferior；

Sub-step 2 is trained, and obtain model file using seller's type data of label using SVM；

Sub-step 3 is participated in the seller that the interaction of dealing man is promoted to selection, is divided using the model file in sub-step 2 Class chooses high-quality seller.

Step 2, the best buy of screening high-quality seller；

In high-quality seller, there are more best buy, the sides of the best buy of screening high-quality seller by each seller Method can specifically include：

Sub-step 1 selects the only commodity of favorable comment and the only poor commodity commented from the merchandising database of high-quality seller, Respectively as positive negative sample；

Sub-step 2, counts the feature of positive negative sample commodity respectively, and Partial Feature carries out Hash acquisition；

Sub-step 3 obtains the high-quality probability of commodity and probability inferior in high-quality businessman using NB Algorithm, and adds Power obtains the factor of merit of commodity；

Sub-step 4, takes out highest N number of factor of merit commodity in each high-quality shop, and respectively obtain the commodity of acquisition Buyer's list.

Step 3, corresponding buyer's list is obtained according to best buy, and filters out high-quality buyer；

There are many features by high-quality buyer, for example, do shopping often, interval time it is short, log in Wang Wang often, log in Main website of Taobao frequently etc..It can specifically include：

Sub-step 1, each buyer's feature-set have corresponding score value, the weight of various buyer's features are extracted, according at least one Kind buyer's feature and corresponding weight, the evaluation score value of each buyer, example are obtained to the corresponding score value weighted calculation of buyer's feature Such as by the registion time of buyer, user gradation, shopping number, nearest one month shopping number, online hours, comment number, It is added after chat messages item number, good friend's quantity, number of contacts weighting and obtains the evaluation score value of buyer.

Sub-step 2, using evaluate score value set interval buyer as high-quality buyer.

It step 4, will be high-quality near active user using the GPS information of the shopping address and active user of high-quality buyer The shopping record of buyer is sent to active user, in the client displaying shopping record of active user；

Step 5, active user obtains the recommendation of commodity according to neighbouring high-quality buyer, is then bought to corresponding seller, at Seller carries out profit-sharing to the high-quality buyer of Recommendations after friendship.

Referring to Fig. 8, the push process schematic of network object in an example of the embodiment of the present application is shown.

Step 1, GPS data of the client timing of high-quality buyer to server update oneself；Server module records excellent The GPS data that the GPS data or transaction address of matter buyer is converted into, and record the shopping items of high-quality buyer and correspond to and sell Family ID；

Step 2, common buyer popped up when client logs in or enter near people's function when pop up；

Step 3, common buyer submits the GPS data of oneself by user end to server and asks neighbouring commodity, clothes Business device calculates common buyer at a distance from each high-quality buyer according to GPS, and returns to corresponding data from the near to the remote according to distance Information；Common buyer can be in client according to from closely browsing commodity respectively to remote sequence；

Step 4, common buyer, if interesting, can contact high-quality buyer and carry out according to the merchandise news of return information Consulting, because distance is closer, it might even be possible to which application is on probation；Common buyer can also contact high-quality seller and link up；

Step 5, common buyer reaches transaction with high-quality seller, and high-quality seller and high-quality buyer reach profit-sharing.

With reference to figure 9, it illustrates the structure diagram according to the pusher of network object of the embodiment of the present application five a kind of, It can specifically include：

Feature seller screening module 501 screens feature buyer for the historical trading based on transaction platform；

Target network object acquisition module 502, for searches with the active user of immediate communication platform with it is associated extremely A few target signature buyer, and obtain at least one target network object that target signature buyer is accessed；

Target network Object Push module 503, for pushing the target to active user by instant communication client Network object.

In this embodiment, it is preferred that the target network object acquisition module includes：

In this embodiment, it is preferred that the feature buyer lookup submodule includes：

In this embodiment, it is preferred that described device further includes：

In this embodiment, it is preferred that further including：

In this embodiment, it is preferred that the immediate communication platform shares same system of account with transaction platform；The dress It sets and further includes：

In this embodiment, it is preferred that described device further includes：

In this embodiment, it is preferred that further including：

In this embodiment, it is preferred that the target network Object Push module includes：

In this embodiment, it is preferred that the transaction entrance carries buyer's mark of target signature buyer；Described device Further include：

In this embodiment, it is preferred that further including：

In this embodiment, it is preferred that the feature seller screening module includes：

Described device further includes：

In this embodiment, it is preferred that the feature buyer screening submodule includes：

In this embodiment, it is preferred that the character network object screening module includes：

In this embodiment, it is preferred that described device further includes：

Described device further includes：

With reference to figure 10, it illustrates the structure diagram according to a kind of instant communication client of the embodiment of the present application seven, tools Body may include：

Push requests to generate module 701, for opening instant communication client according to active user or entering to push function The access operation of mouth generates the push request of network object；

Request sending module 702, for sending the push request to server-side；

Target network object receiving module 703, the target network object for receiving server-side push, the target network For object to have the network object that associated target signature buyer is accessed with active user, it is flat that the feature buyer is based on transaction The historical trading screening of platform obtains.

According to a kind of structure diagram of information interaction device of the embodiment of the present application eight, can specifically include：

Displaying control, push function entrance；

Wherein, the displaying control is for showing functional entrance；

With reference to figure 11, it illustrates the structural frames according to the pusher of network object of the embodiment of the present application nine a kind of Figure, can specifically include：

Buyer's module 801 is screened, for screening feature buyer based on historical trading；

Object acquisition module 802, it is special for searching at least one target being in set distance range with active user Buyer is levied, and obtains at least one target network object that target signature buyer is accessed；

Object Push module 803, for pushing the target network object to active user.

With reference to figure 12, it illustrates the structural frames according to the supplying system of network object of the embodiment of the present application ten a kind of Figure, can specifically include：

Transaction platform 901, immediate communication platform 902, instant communication client 903；

Immediate communication platform 902 screens feature buyer for the historical trading based on transaction platform 901；And search with The active user of immediate communication platform 902 has associated at least one target signature buyer, and obtains target signature buyer institute At least one target network object accessed；

Instant communication client 903, for receive the immediate communication platform 902 transmission target network object and carry out Displaying.

Since described device embodiment essentially corresponds to earlier figures 1, Fig. 2, Fig. 3, Fig. 4 and embodiment of the method shown in fig. 5, Therefore not detailed place in the description of the present embodiment, the related description in previous embodiment is may refer to, is not just repeated herein.

Algorithm and display be not inherently related to any certain computer, virtual system or miscellaneous equipment provided herein. Various general-purpose systems can also be used together with teaching based on this.As described above, it constructs required by this kind of system Structure be obvious.In addition, the application is not also directed to any certain programmed language.It should be understood that can utilize various Programming language realizes present context described here, and the description done above to language-specific is to disclose this Shen Preferred forms please.

In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the application Example can be put into practice without these specific details.In some instances, well known method, structure is not been shown in detail And technology, so as not to obscure the understanding of this description.

Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of each application aspect, Above in the description of the exemplary embodiment of the application, each feature of the application is grouped together into single implementation sometimes In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention：It is i.e. required to protect Shield this application claims the more features of feature than being expressly recited in each claim.More precisely, as following Claims reflect as, all features less than single embodiment disclosed above are in terms of application.Therefore, Thus the claims for following specific implementation mode are expressly incorporated in the specific implementation mode, wherein each claim itself All as the separate embodiments of the application.

Those skilled in the art, which are appreciated that, to carry out adaptively the module in the equipment in embodiment Change and they are arranged in the one or more equipment different from the embodiment.It can be the module or list in embodiment Member or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it may be used any Combination is disclosed to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so to appoint Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including adjoint power Profit requires, abstract and attached drawing) disclosed in each feature can be by providing the alternative features of identical, equivalent or similar purpose come generation It replaces.

In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments In included certain features rather than other feature, but the combination of the feature of different embodiments means to be in the application's Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed One of meaning mode can use in any combination.

The all parts embodiment of the application can be with hardware realization, or to run on one or more processors Software module realize, or realized with combination thereof.It will be understood by those of skill in the art that can use in practice Microprocessor or digital signal processor (DSP) realize that the server based on data analysis according to the embodiment of the present application enters Invade some or all functions of some or all components in identification equipment.The application is also implemented as executing this In described method some or all equipment or program of device (for example, computer program and computer program Product).It is such realize the application program can may be stored on the computer-readable medium, can either have there are one or it is more The form of a signal.Such signal can be downloaded from internet website and be obtained, either provide on carrier signal or with Any other form provides.

The application is limited it should be noted that above-described embodiment illustrates rather than the application, and ability Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims, Any reference mark between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such Element.The application can be by means of including the hardware of several different elements and being come by means of properly programmed computer real It is existing.In the unit claims listing several devices, several in these devices can be by the same hardware branch To embody.The use of word first, second, and third does not indicate that any sequence.These words can be explained and be run after fame Claim.

Claims

1. a kind of method for pushing of network object, which is characterized in that including：

Historical trading based on transaction platform screens feature buyer, and the transaction platform is directed to network for providing buyer and seller The transactional services of object；

Search has associated at least one target signature buyer with the active user of immediate communication platform, and obtains target signature At least one target network object that buyer is accessed, the immediate communication platform are supported to hand over immediately based on internet between user Flow message；

2. the method as described in claim 1, which is characterized in that the active user of the lookup and immediate communication platform, which has, closes Connection at least one feature buyer include：

3. method as claimed in claim 2, which is characterized in that the active user of the lookup and immediate communication platform, which is in, to be set At least one feature buyer within the scope of set a distance includes：

4. method as claimed in claim 2, which is characterized in that pushed to active user by instant communication client described Before the target network object, the method further includes：

It is sequentially sorted according to the distance of character pair buyer and active user according to by the target network object.

5. method as claimed in claim 3, which is characterized in that the method further includes：

The request that network object is pushed to active user is received, the request carries the geographical location information of the active user, The request is generated when active user opens instant communication client or accesses push function entrance.

6. method as claimed in claim 3, which is characterized in that further include：

The geographical location information that feature buyer periodically submits is received, or the transaction address information of the feature buyer is converted to ground Manage location information.

7. the method as described in claim 1, which is characterized in that the immediate communication platform shares same account with transaction platform System；

It is described by instant communication client to active user push the target network object when, the method further includes：

The target seller of the target network object is pushed to the active user.

8. the method for claim 7, which is characterized in that accessed in the acquisition target signature buyer at least one When target network object, the method further includes：

Obtain the target seller that the target network object is provided.

9. the method for claim 7, which is characterized in that further include：

The communication message sent to the target seller by instant communication client is received, and the communication message is forwarded to The instant communication client of target signature buyer.

10. the method as described in claim 1, which is characterized in that described to be pushed to active user by instant communication client The target network object includes：

The push page for target network object is sent to the instant communication client of active user, is wrapped in the push page Transaction entrance is included, the transaction entrance is operated to the purchase for coping with target network object or operation on probation.

11. method as claimed in claim 10, which is characterized in that the transaction entrance carries the buyer of target signature buyer Mark；

The method further includes：

Buyer's mark that transaction request carries is extracted, the avail data of this corresponding transaction request, which is increased to buyer, identifies label Target signature buyer account data.

12. method as claimed in claim 11, which is characterized in that further include：

13. the method as described in claim 1, which is characterized in that the historical trading screening feature based on transaction platform is bought Family include：

From character network object in the corresponding all buyers of historical trading, bought according at least one buyer's attribute selection feature Family；

14. method as claimed in claim 13, which is characterized in that described according at least one buyer's attribute selection feature buyer Including：

15. method as claimed in claim 13, which is characterized in that at least one of the network object according to historical trading Object properties screen character network object：

Extraction section network object obtains the object classification label of network of samples object as network of samples object, described right As tag along sort includes character network object and non-character network object；

The object classification label and object properties of statistical sample network object generate and predict object classification label according to object properties Prediction data set；

Using the object classification label of each network object of prediction data set prediction history transaction, and according to object classification Tag extraction character network object.

16. method as claimed in claim 13, which is characterized in that at least the one of the network object according to historical trading Before kind object properties screening character network object, the method further includes：

Extract the transaction feature data of each network object from various dimensions, the transaction feature data include trading activity data and mutually At least one of dynamic behavioral data；

Judge the numerical intervals of the transaction feature attribution data, and search the object properties that the numerical intervals correspond to instruction, The numerical intervals divide to obtain according to the transaction feature data of all-network object.

17. method as claimed in claim 13, which is characterized in that at least one of the network object according to historical trading Object properties screen character network object：

From feature seller in the corresponding all-network object of historical trading, character network is screened according at least one object properties Object；

Before at least one object properties screening character network object of the network object according to historical trading, institute The method of stating further includes：

18. method as claimed in claim 17, which is characterized in that described according at least one seller's attribute selection feature seller Including：

Extraction section seller obtains seller's tag along sort of sample seller as sample seller, seller's tag along sort packet Include feature seller and non-feature sellers；

Seller's tag along sort of statistical sample seller and sell it is attribute, generate according to the pre- of seller's attribute forecast seller's tag along sort Measured data set；

Seller's tag along sort of each seller is predicted using the prediction data set, and feature is extracted according to seller's tag along sort and is sold Family.

19. method as claimed in claim 17, which is characterized in that described according at least one seller's attribute selection feature seller Including：

20. a kind of method for pushing of network object, which is characterized in that including：

Instant communication client is opened according to active user or pushing away for network object is generated to the access operation of push function entrance Send request；

The push request is sent to server-side；

The target network object of server-side push is received, the target network object is to have associated target special with active user The network object that sign buyer is accessed, historical trading screening of the feature buyer based on transaction platform obtain.

21. a kind of information interacting method, which is characterized in that including：

Show push function entrance；

Receive the access operation to push function entrance；

The push request of network object is generated, and is sent to server-side；

The target network object of server-side push is received, the target network object is to have associated target special with active user The network object that sign buyer is accessed, historical trading screening of the feature buyer based on transaction platform obtain；

22. a kind of method for pushing of network object, which is characterized in that including：

Feature buyer is screened based on historical trading；

It searches and is at least one target signature buyer in set distance range with active user, and obtain target signature buyer At least one target network object accessed；

The target network object is pushed to active user.

23. a kind of pusher of network object, which is characterized in that including：

Target network object acquisition module has associated at least one mesh for searching with the active user of immediate communication platform Feature buyer is marked, and obtains at least one target network object that target signature buyer is accessed；

Target network Object Push module, for pushing the target network pair to active user by instant communication client As.

24. a kind of instant communication client, which is characterized in that including：

Push requests to generate module, for according to active user's unlatching instant communication client or the access to push function entrance Operation generates the push request of network object；

Request sending module, for sending the push request to server-side；

Target network object receiving module, the target network object for receiving server-side push, the target network object are There is the network object that associated target signature buyer is accessed, feature buyer the going through based on transaction platform with active user History transaction screening obtains.

25. a kind of information interaction device, which is characterized in that including：

Displaying control, push function entrance；

Wherein, the displaying control is for showing functional entrance；

The push that described information interactive device is used to generate network object according to access operation is asked, and is sent to server-side；It connects The target network object of server-side push is received, the target network object is to have associated target signature buyer with active user The network object accessed, historical trading screening of the feature buyer based on transaction platform obtain；

The displaying control is additionally operable to show the target network object and provides the target seller of the target network object.

26. a kind of pusher of network object, which is characterized in that including：

Object acquisition module, for searching at least one target signature buyer being in active user in set distance range, And obtain at least one target network object that target signature buyer is accessed；

Object Push module, for pushing the target network object to active user.

27. a kind of supplying system of network object, which is characterized in that including：

The immediate communication platform screens feature buyer for the historical trading based on transaction platform；And search and Instant Messenger The active user for interrogating platform has associated at least one target signature buyer, and obtains target signature buyer and accessed at least One target network object；

The instant communication client is used to receive the target network object that the immediate communication platform is sent and is shown.