CN108346090A - Method of Commodity Recommendation and device - Google Patents
Method of Commodity Recommendation and device Download PDFInfo
- Publication number
- CN108346090A CN108346090A CN201810177406.5A CN201810177406A CN108346090A CN 108346090 A CN108346090 A CN 108346090A CN 201810177406 A CN201810177406 A CN 201810177406A CN 108346090 A CN108346090 A CN 108346090A
- Authority
- CN
- China
- Prior art keywords
- user
- goods
- lipuid
- record
- scoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Embodiment of the present invention is related to field of information processing, discloses a kind of Method of Commodity Recommendation, including regularly updates the purchaser record and scoring record of each user in store;Detecting the purchaser record that each user is obtained when the lipuid goods information promoted and scoring record;It is recorded according to the purchaser record of each user and scoring, judges whether the PUSH message for sending lipuid goods to user, and PUSH message is sent when sentencing directional user's transmission PUSH message, then to user.Embodiment of the present invention also discloses a kind of device for recommending the commodity, including:Logging modle, acquisition module, judgment module and pushing module.The Method of Commodity Recommendation and the device for recommending the commodity that embodiment of the present invention is provided have targetedly to user's Recommendations, avoid user from generating the advantages of disliking while improving marketing conversion ratio.
Description
Technical field
The present embodiments relate to field of information processing, more particularly to the recommendation method and device of a kind of commodity.
Background technology
With Internet technology continuous development with it is universal, online shopping mall is also more and more.In order to attract more consumption
The Pages Design of person, online shopping mall are also more and more abundant, and the effect that different commodity are shown is also more and more personalized.With competing
The various marketing activities of the aggravation striven, online shopping mall are also more and more, in order to make more users know that marketing activity is sold
Commodity, the commodity that marketing activity is sold, which are pushed to user, to be necessary.
Existing Method of Commodity Recommendation is directly straight by modes such as movable prefecture, push note message or system messages
It connects guiding or recommends user.
However, it was found by the inventors of the present invention that existing Method of Commodity Recommendation, there are problems that blindly pushing, push battalion
Pin commodity do not have specific aim, and marketing conversion ratio is relatively low, i.e., no matter whether user likes Recommendations or whether rigid before activity
Recommendations just were bought, all can send recommendation information to user, more invalid rubbish marketing letter can be pushed to certain user
Breath allows user to generate dislike.
Invention content
Embodiment of the present invention is designed to provide a kind of Method of Commodity Recommendation and device, is targetedly pushed away to user
Commodity are recommended, avoid user from generating dislike while improving marketing conversion ratio.
In order to solve the above technical problems, embodiments of the present invention provide a kind of Method of Commodity Recommendation, including following step
Suddenly:The purchaser record and scoring for regularly updating each user in store record;It is detecting when the lipuid goods information promoted, is obtaining
The purchaser record and scoring for taking family record;It is recorded according to the purchaser record of user and scoring, judges whether to send to user and live
The PUSH message of dynamic commodity, and send PUSH message when sentencing directional user's transmission PUSH message, then to user.
Embodiments of the present invention additionally provide a kind of device for recommending the commodity, including:Logging modle, for regularly updating quotient
The purchaser record of each user in city and scoring record;Acquisition module, for detecting when the lipuid goods information promoted, from
The purchaser record and scoring that user is obtained in logging modle record;Judgment module, the user's for being used to be obtained according to acquisition module
Purchaser record and scoring record, and judge whether the PUSH message that lipuid goods are sent to user;Pushing module, for judging mould
When block sentences the PUSH message of directional user's transmission lipuid goods, PUSH message is sent to user.
Embodiment of the present invention in terms of existing technologies, lipuid goods information to be promoted, quotient is detected in store
When the Platform Server in city needs to send the PUSH message of lipuid goods to user, the purchaser record of each user can be obtained respectively
It records with scoring, and is recorded according to the purchaser record of each user of acquisition and scoring, judge whether to be sent to lipuid goods
PUSH message, i.e., by the purchaser record of each user and scoring record, select be suitble to recommend this lipuid goods user, only
Have when judgement is sent to the PUSH message of lipuid goods, can just be sent to the PUSH message of lipuid goods;I.e. only to suitable
The PUSH message for recommending the user of this lipuid goods to send this lipuid goods is closed, those are not suitable for recommend this lipuid goods
User will not then be sent to the PUSH message of lipuid goods, due to being suitble to that the user of this lipuid goods is recommended to buy this activity
Commodity probability, which can be much larger than, is not suitable for that the user of this lipuid goods is recommended to buy this lipuid goods probability, only recommends this work to suitable
The user of dynamic commodity sends the PUSH message of this lipuid goods, can effectively promote the marketing conversion ratio of lipuid goods;Meanwhile prohibiting
Only to the unsuitable PUSH message for recommending the user of this lipuid goods to send lipuid goods, it is possible to reduce the rubbish that user receives disappears
The quantity of breath prevents user since excessive rubbish message generates dislike, promotes the likability of user.Further, due to quotient
The Platform Server in city can regularly update the purchaser record of each user in store and scoring records, it is ensured that selection is suitble to recommend
The real-time and accuracy of foundation when the user of this lipuid goods.
In addition, regularly updating the purchaser record and scoring record of each user in store, specifically include:With the first default week
Phase periodically inquires the goods number of the purchased item of each user and the scoring record of purchased item;After inquiring each time,
According to the quantity purchase of the purchased item of user, quantity sequence, and commenting according to purchased item are carried out to purchased item
Member record carries out marking and queuing to purchased item;Wherein, the more purchased item of quantity purchase, in quantity sequence
Ranking is more forward;The higher purchased item of scoring, the ranking in marking and queuing are more forward;With the second predetermined period, periodically
According to the sequence of the quantity of each user and marking and queuing, the purchaser record and scoring that update each user record;Wherein, the purchase of user
Record includes the goods number of the preceding X purchased item in the quantity sequence of user, and the scoring record of user includes user's
The goods number of preceding Y purchased item in marking and queuing;X and Y is the natural number more than or equal to 1.User's purchase
The type and quantity of commodity are more complicated, choose purchaser record of the preceding X most commodity of user's purchase as user, can be with
While effective simplified the step for judging whether to send the recommendation message of lipuid goods to user according to purchaser record, protect
Demonstrate,prove the reference significance of purchaser record;Similarly, the commodity that user is evaluated are also more complicated, choose the highest preceding Y of user's scoring
The commodity of name are recorded as the scoring of user, can be judged whether to user's transmission activity according to scoring record in effective simplify
While the step for recommendation message of commodity, ensure the reference significance of scoring record.
In addition, being recorded according to the purchaser record of user and scoring, judge whether that the push that lipuid goods are sent to user disappears
Breath, specifically includes:In the purchaser record or the scoring record that detect user, if the goods number including lipuid goods;If including
The goods number of lipuid goods then sentences the PUSH message that directional user sends lipuid goods.Due to the purchase that purchaser record is user
Buy the goods number of the preceding X purchased item in quantity sequence, the i.e. preceding X purchased item of user's quantity purchase at most
The purchaser record of goods number, user includes the goods number of lipuid goods, indicates that user's purchase lipuid goods are more, to purchase
Buy record include lipuid goods goods number user send lipuid goods PUSH message, user buy probability compared with
Greatly, marketing conversion ratio can effectively be promoted;Similarly, it has been bought due to scoring preceding Y be recorded as in the marking and queuing of user
The goods number of commodity, i.e. user score the goods number of highest preceding Y purchased item, and the scoring record of user includes
The goods number of lipuid goods indicates that user is higher to the degree of liking of lipuid goods, includes lipuid goods to scoring record
The user of goods number sends the PUSH message of lipuid goods, and the probability of user's purchase is larger, can further promote marketing
Conversion ratio.
In addition, being recorded according to the purchaser record of user and scoring, judge whether that the push that lipuid goods are sent to user disappears
Before breath, further include:In the goods number for the commodity that detection user has bought in preset duration, if there are lipuid goods
Goods number;If there are the goods number of lipuid goods, forbid sending PUSH message to user;If there is no lipuid goods
Goods number is then executed and is recorded according to the purchaser record and scoring of user, judges whether to send pushing away for lipuid goods to user again
Send message.User, there are the goods number of lipuid goods, illustrates to use in the goods number of the default often interior commodity bought
Lipuid goods had been bought at family recently, and the probability that user buys again is smaller, forbade sending PUSH message to user, can be into one
Step reduces the rubbish message for being sent to user, prevents user from generating dislike.
In addition, the scoring record of user further includes:The commodity of rear Z purchased item in the marking and queuing of user are compiled
Number, Z is the natural number more than or equal to 1;It is recorded, is judged whether to user's transmission activity according to the purchaser record of user and scoring
The PUSH message of commodity, specifically includes:In the purchaser record for detecting user, if the goods number including lipuid goods;If
Goods number including lipuid goods then detects the goods number of lipuid goods, if is present in the commodity of Z purchased item
In number;If the goods number of lipuid goods, it is present in the goods number of Z purchased item, then judges not send out to user
Send the PUSH message of lipuid goods;If the goods number of lipuid goods is not present in the goods number of Z purchased item,
Then sentence the PUSH message that directional user sends lipuid goods.Scoring record further includes that rear Z in the marking and queuing of user have purchased
The goods number of commodity is bought, Z is a after the marking and queuing of ranked users, indicates that user is relatively low to the degree of liking of such commodity, activity
The goods number of commodity is present in the goods number of Z purchased item, does not disappear to the push for stating user's transmission lipuid goods
Breath can be further reduced the junk information sent to user, prevent to avoid the recommendation information for the commodity not liked is sent to user
Only user generates dislike.
Description of the drawings
One or more embodiments are illustrated by the picture in corresponding attached drawing, these exemplary theorys
The bright restriction not constituted to embodiment, the element with same reference numbers label is expressed as similar element in attached drawing, removes
Non- to have special statement, composition does not limit the figure in attached drawing.
Fig. 1 is the program flow diagram of the Method of Commodity Recommendation provided according to first embodiment of the invention;
Fig. 2 is the program flow diagram of the Method of Commodity Recommendation provided according to second embodiment of the invention;
Fig. 3 is the program flow diagram of the Method of Commodity Recommendation provided according to third embodiment of the invention;
Fig. 4 is the structural schematic diagram of the device for recommending the commodity provided according to four embodiment of the invention;
Fig. 5 is the structural schematic diagram of the device for recommending the commodity provided according to fifth embodiment of the invention.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Each embodiment be explained in detail.However, it will be understood by those skilled in the art that in each embodiment party of the present invention
In formula, many technical details are proposed in order to make reader more fully understand the present invention.But even if without these technical details
And various changes and modifications based on the following respective embodiments, it can also realize claimed technical solution of the invention.
The first embodiment of the present invention is related to a kind of Method of Commodity Recommendation, the core of present embodiment is:Periodically more
The purchaser record of each user in the new store and scoring record;It is detecting when the lipuid goods information promoted, is obtaining institute
The purchaser record and scoring for stating user record;It is recorded, is judged whether to the use according to the purchaser record of the user and scoring
Family sends the PUSH message of the lipuid goods, and when judgement sends the PUSH message to the user, then to the use
Family sends the PUSH message.The implementation detail of the Method of Commodity Recommendation of present embodiment is specifically described below, with
Lower content only for convenience of the implementation detail provided is understood, not implements the necessary of this programme.
The first embodiment of the present invention is related to a kind of Method of Commodity Recommendation.Detailed process is as shown in Figure 1, include:
Step S101:The purchaser record and scoring for regularly updating each user in store record.
Specifically, user does shopping in store every time, the Platform Server in store can all record the quotient of user's purchase
The number of product and the quantity of purchase;The Platform Server in store can inquire user first at interval of the first predetermined period
The goods number for the commodity bought in predetermined period and the quantity of purchase, and bought in the first predetermined period according to user
Quantity, the commodity amount sequence of update user's purchase, the more purchased item of quantity purchase, the ranking in quantity sequence are got over
It is forward;Meanwhile user every time scores to the commodity in store, the Platform Server in store can all record user and score
Commodity goods number and scoring numerical value, and the commodity to have been scored to user according to the numerical value of scoring carry out marking and queuing,
The higher commodity of scoring, then the ranking in this marking and queuing are more forward.
After the quantity sequence for establishing the commodity that user has bought and marking and queuing, at interval of the second predetermined period, more
The purchaser record of new user and scoring record, wherein the purchaser record of user are updated, specifically, according to last update
The quantity for the commodity that user has bought sorts, and chooses the purchased item of X before ranking in quantity sequence, and obtains the quotient of its commodity
Product are numbered, using the goods number of the commodity of acquisition as the purchaser record of newest user.The scoring record of user is updated, specifically
According to the marking and queuing of the user of last update, to choose the scoring commodity of Y before ranking in marking and queuing, and obtain
The goods number of its commodity is recorded the goods number of the commodity of acquisition as the scoring of newest user.
It is understood that the first predetermined period may be the same or different with the second predetermined period, the specific time
It can be chosen according to actual needs;X and Y can be chosen according to actual needs, X and Y values are bigger, then what is pushed is general
Rate is bigger, X and Y values are smaller, then the user pushed is more accurate.Such as the first predetermined period can be one day, second is default
Period can be one week, i.e. the Platform Server in store will every other day count the quotient for the commodity that each user buys in one day
The numerical value for the quantity and scoring that product are numbered and bought, and update quantity purchase sequence and the marking and queuing of user;At interval of one
Week all the goods number of first X and preceding Y of commodity can be chosen respectively according to the sequence of newest quantity purchase and marking and queuing,
It is recorded as the newest purchaser record of user and scoring.
Step S102:Detect that the purchaser record that the user is obtained when the lipuid goods information promoted and scoring are remembered
Record.
Specifically, it when store detects lipuid goods information to be promoted, needs to send pushing away for lipuid goods to user
When sending message, the Platform Server in store can obtain the purchaser record of user and scoring records, i.e. the quantity purchase sequence of user
Before middle ranking in the goods number of the purchased item of X and the marking and queuing of user before ranking the goods number of the commodity of Y and
The goods number of the commodity of Z after ranking.
Step S103:It is recorded according to the purchaser record of user and scoring, judges whether to send pushing away for lipuid goods to user
Message is sent, when sentencing directional user and sending the PUSH message of lipuid goods, executes step S104, it is no to then follow the steps S105.
Specifically, after the purchaser record and scoring record for obtaining user, the Platform Server in store can be by movable quotient
The goods number of commodity in the goods number and purchaser record of product is compared one by one, judge in the purchaser record of user whether
There are the goods numbers of lipuid goods, if there are the goods numbers of lipuid goods, i.e. user to buy in the purchaser record of user
Lipuid goods, and quantity purchase is more, then illustrates that user has purchasing demand to lipuid goods, and lipuid goods are sent to such user
PUSH message, user is more likely to purchase lipuid goods;Meanwhile the Platform Server in store can compile the commodity of lipuid goods
The goods number of commodity in number being recorded with scoring is compared one by one, is judged in the scoring record of user with the presence or absence of activity quotient
The goods number of product illustrates user to lipuid goods if there are the goods numbers of lipuid goods in the scoring record of user
Degree of liking is higher, and the PUSH message of lipuid goods is sent to such user, and user is also more likely to purchase lipuid goods.
Step S104:The PUSH message of lipuid goods is sent to user.
Specifically, when judging to include lipuid goods in purchaser record or scoring record, the Platform Server in store is
The PUSH message that lipuid goods can be sent to user recommends lipuid goods to user.
Step S105:Forbid sending the PUSH message of lipuid goods to user.
Specifically, when judging not including lipuid goods in purchaser record and scoring record, the platform service in store
Device will not send the PUSH message of lipuid goods to user, not recommend lipuid goods to user.
It should be noted that can also include being arranged in user's marking and queuing in the scoring record of user in practical applications
The goods number of the purchased item of Z after name.
When the goods number for including lipuid goods in the purchaser record that the Platform Server in store detects user, i.e. user
Bought lipuid goods;At this point, whether the goods number that the Platform Server in store can detect lipuid goods is remembered in above-mentioned scoring
In the goods number of rear Z purchased item in record, if, illustrate although user bought the lipuid goods, but it is right
The degree of liking of lipuid goods is relatively low, then judges the PUSH message that lipuid goods are not sent to user;If not existing, no matter movable quotient
Product all send the PUSH message of lipuid goods whether in the preceding Y goods numbers bought during scoring records to user.
It is understood that Z is the natural number more than or equal to 1;It is understood that Z can according to actual needs into
Row is chosen, and the value of Z is bigger, then the user pushed is more accurate, and Z values are smaller, then judges that the probability of push is bigger.Certainly, Y and
The value of Z should not make preceding Y and latter Z generation intersections of the marking and queuing of user.
It compared with prior art, can be according to the purchase of user when needing to send the PUSH message of lipuid goods to user
Record and scoring record judge whether to send the PUSH message of lipuid goods to user, are suitble to send lipuid goods to choose
The user of PUSH message only sends the PUSH message of lipuid goods to the user of the suitable PUSH message for sending lipuid goods, right
In the user that those are not suitable for recommending this lipuid goods, then it will not be sent to the PUSH message of lipuid goods, due to being suitble to push away
Recommend this lipuid goods user buy this lipuid goods probability can be much larger than be not suitable for recommend this lipuid goods user buy this
Lipuid goods probability can be carried effectively only to the suitable PUSH message for recommending the user of this lipuid goods to send this lipuid goods
Rise the marketing conversion ratio of lipuid goods;Meanwhile not to the unsuitable push for recommending the user of this lipuid goods to send lipuid goods
Message, it is possible to reduce the rubbish message that user receives prevents user since excessive rubbish message generates dislike, promotes user's
Likability.Further, since the Platform Server in store can regularly update the purchaser record of each user in store and scoring is remembered
Record, it is ensured that the real-time and accuracy of foundation when selection is suitble to recommend the user of this lipuid goods.
Second embodiment of the present invention is related to a kind of Method of Commodity Recommendation.Second embodiment is to first embodiment
Be further improved, mainly the improvement is that:In second embodiment of the invention, according to the purchaser record of user and commenting
Member record judges whether before the PUSH message that user sends lipuid goods, also needs to buy lipuid goods according to user's last time
Time, judge whether to forbid the recommendation message for sending lipuid goods to user, be as follows:
Step S201:The purchaser record and scoring for regularly updating each user in store record.
Step S202:Detect that the purchaser record that the user is obtained when the lipuid goods information promoted and scoring are remembered
Record.
Step S203:Judge whether user bought lipuid goods in preset duration, if user purchases in purchase duration
Lipuid goods were bought, S206 is thened follow the steps, it is no to then follow the steps S204.
Specifically, user bought lipuid goods in preset duration, user has lipuid goods the possibility of demand
It is relatively low, to user send lipuid goods recommendation message, user buy lipuid goods possibility it is smaller, marketing conversion ratio compared with
Low, the Platform Server in store then judges not send the PUSH message of lipuid goods to user, executes step S206;Otherwise it executes
Step S204 further judges whether the PUSH message that lipuid goods are sent to user.
Step S204:The PUSH message for judging whether to send lipuid goods to user, when sentencing directional user transmission activity quotient
When the PUSH message of product, step S205 is executed, it is no to then follow the steps S206.
Step S205:The PUSH message of lipuid goods is sent to user.
Step S206:The PUSH message of lipuid goods is not sent to user.
Due in present embodiment step S201, S202 and step S204 to step S206 and first embodiment in
Step S101 it is roughly the same to step S105, it is intended to the purchaser record of regular thinner user and scoring record, and according to purchase
Record and scoring record, judge whether the PUSH message that lipuid goods are sent to user, details are not described herein.
Compared with prior art, second embodiment of the present invention is needing the PUSH message to user's transmission lipuid goods
When, it can judge user is presetting in often whether bought lipuid goods, if user bought lipuid goods in preset duration,
Illustrate that demand of the user to lipuid goods is smaller, then forbids the PUSH message for sending lipuid goods to user, to promote store
Conversion ratio of marketing executed the purchaser record according to user and scoring if user does not buy lipuid goods in preset duration
Record judges whether to send the PUSH message of lipuid goods to user, to choose the PUSH message for being suitble to send lipuid goods
User only sends the PUSH message of lipuid goods to the user of the suitable PUSH message for sending lipuid goods, for those discomforts
The user for recommending this lipuid goods is closed, then will not be sent to the PUSH message of lipuid goods, due to being suitble to recommend this movable quotient
The user of product buys this lipuid goods probability and can be much larger than the user that is not suitable for recommending this lipuid goods to buy this lipuid goods several
Rate can effectively promote lipuid goods only to the suitable PUSH message for recommending the user of this lipuid goods to send this lipuid goods
Marketing conversion ratio;Meanwhile it can not subtract to the unsuitable PUSH message for recommending the user of this lipuid goods to send lipuid goods
The rubbish message that few user receives prevents user since excessive rubbish message generates dislike, promotes the likability of user.Into one
Step, since the Platform Server in store can regularly update the purchaser record and scoring record of each user in store, it is ensured that
The real-time and accuracy of foundation when selection is suitble to recommend the user of this lipuid goods.
Third embodiment of the present invention is related to a kind of Method of Commodity Recommendation, and third embodiment and second embodiment are big
Cause identical, the main distinction is, in this second embodiment, the purchaser record of user and scoring be recorded as user buy commodity and
The goods number of the commodity to score, and in the third embodiment, the purchaser record of user and scoring are recorded as user's purchase
The type of merchandize for the commodity bought commodity and scored is numbered;Detailed process is as shown in figure 3, include the following steps:
Step S301:The purchaser record and scoring for regularly updating each user in store record.
Specifically, user does shopping in store every time, and the Platform Server in store can all record the quotient of user's purchase
The type of merchandize of product is numbered and the quantity of purchase;The Platform Server in store can inquire use at interval of the first predetermined period
The type of merchandize for the commodity that family is bought in the first predetermined period is numbered and the quantity of purchase, and default first according to user
The quantity bought in period, the same type commodity amount sequence of update user's purchase, the more purchased item of quantity purchase,
Ranking in quantity sequence is more forward;Meanwhile user every time scores to the commodity in store, the Platform Server in store is all
The type of merchandize for the commodity that user scores and the numerical value of scoring can be recorded, and scored user according to the numerical value of scoring
It is more forward that type of merchandize carries out marking and queuing, the higher type of merchandize of scoring, then the ranking in this marking and queuing.
After the quantity sequence for establishing the type of merchandize that user has bought and marking and queuing, at interval of the second default week
Phase, the purchaser record and scoring for updating user record, wherein the purchaser record of user are updated, specifically, more according to the last time
The quantity for the type of merchandize that new user has bought sorts, and chooses the purchased item type of X before ranking in quantity sequence, and obtains
It takes its type of merchandize to number, the type of merchandize of acquisition is numbered into the purchaser record as newest user.Update the scoring of user
Record chooses the scoring quotient of Y before ranking in marking and queuing specifically, according to the marking and queuing of the user of last update
Kind class, and its type of merchandize number is obtained, the type of merchandize number of acquisition is recorded as the scoring of newest user.
It is understood that the first predetermined period may be the same or different with the second predetermined period, the specific time
It can be chosen according to actual needs;X and Y can be chosen according to actual needs, X and Y values are bigger, then the use pushed
Family is more, X and Y values are smaller, then the user pushed is more accurate.
Step S302:Detect that the purchaser record that the user is obtained when the lipuid goods information promoted and scoring are remembered
Record.
Specifically, it when store detects lipuid goods information to be promoted, needs to send pushing away for lipuid goods to user
When sending message, the Platform Server in store can obtain the purchaser record of user and scoring records, i.e. the quantity purchase sequence of user
Before middle ranking in the purchased item type number of X and the marking and queuing of user before ranking Y type of merchandize number and ranking
The type of merchandize number of Z afterwards.
Step S303:Judge whether user bought the congener commodity of lipuid goods in preset duration, if user exists
The congener commodity of lipuid goods were bought in purchase duration, then follow the steps S306, it is no to then follow the steps S304.
Specifically, user bought the congener commodity of lipuid goods in preset duration, user has lipuid goods
The possibility of demand is relatively low, and the recommendation message of lipuid goods is sent to user, and the possibility that user buys lipuid goods is smaller, battalion
Pin conversion ratio is relatively low, and the Platform Server in store then judges not send the PUSH message of lipuid goods to user, executes step
S206;It is no to then follow the steps S204, further judge whether the PUSH message that lipuid goods are sent to user.
Step S304:It is recorded according to the purchaser record of user and scoring, judges whether to send pushing away for lipuid goods to user
Message is sent, when sentencing directional user and sending the PUSH message of lipuid goods, executes step S305, it is no to then follow the steps S306.
Specifically, after the purchaser record and scoring record for obtaining user, the Platform Server in store can be by movable quotient
The type of merchandize number of product is compared one by one with the type of merchandize number of the commodity in purchaser record, judges the purchase note of user
It is numbered with the presence or absence of the type of merchandize of lipuid goods in record, if there are the type of merchandize of lipuid goods volumes in the purchaser record of user
Number, i.e., user bought the congener commodity of lipuid goods, and quantity purchase is more, then illustrated that user has purchase to lipuid goods
Demand, the PUSH message of lipuid goods is sent to such user, and user is more likely to purchase lipuid goods;Meanwhile store is flat
Platform server can be a pair of by the type of merchandize of lipuid goods number and the type of merchandize number progress one of the commodity in the record that scores
Than judging that the type of merchandize in the scoring record of user with the presence or absence of lipuid goods is numbered, if existing in the scoring record of user
The type of merchandize of lipuid goods is numbered, and judges that the type of merchandize number of lipuid goods is the quotient for belonging to Y before ranking in marking and queuing
The type of merchandize of Z after kind class or ranking then illustrates user couple if belonging to the type of merchandize of Y before ranking in marking and queuing
The degree of liking of the commodity of lipuid goods same type is higher, and the PUSH message of lipuid goods is sent to such user, and user also more has
It is likely to purchase lipuid goods;If belonging to the type of merchandize of Z after ranking in marking and queuing, then illustrate that user is of the same race to lipuid goods
The degree of liking of class commodity is relatively low, to such user send lipuid goods recommendation message, user buy lipuid goods probability compared with
Small, bigger is likely to become rubbish message, and user is made to generate dislike, then the Platform Server in store can be forbidden sending to such user
The recommendation message of lipuid goods.
Step S305:The PUSH message of lipuid goods is sent to user.
Step S306:The PUSH message of lipuid goods is not sent to user.
Since step S305, S306 are roughly the same with step S205, S206 of second embodiment, it is intended to according to step
The judgement of S304 is as a result, sending the PUSH message of lipuid goods to user from the background server in store or not sending work to user
The PUSH message of dynamic commodity, details are not described herein.
Compared with prior art, third embodiment of the present invention is needing the PUSH message to user's transmission lipuid goods
When, it can judge user is presetting in often whether bought the congener commodity of lipuid goods, if user purchases in preset duration
The congener commodity of lipuid goods were bought, illustrate that demand of the user to lipuid goods is smaller, then were forbidden to user transmission activity quotient
The PUSH message of product, to promote the marketing conversion ratio in store, if user did not bought the same type of lipuid goods in preset duration
Commodity, then execute and judge whether that the push for sending lipuid goods to user disappears according to the purchaser record of user and scoring record
Breath, to choose the user for the PUSH message for being suitble to send lipuid goods, only to the suitable PUSH message for sending lipuid goods
User sends the PUSH message of lipuid goods, those are not suitable for recommending with the user of this lipuid goods, then will not be sent to
The PUSH message of lipuid goods can be much larger than discomfort due to being suitble to the user of this lipuid goods of recommendation to buy this lipuid goods probability
It closes and the user of this lipuid goods is recommended to buy this lipuid goods probability, only recommend the user of this lipuid goods to send this work to suitable
The PUSH message of dynamic commodity, can effectively promote the marketing conversion ratio of lipuid goods;Meanwhile this movable quotient is not recommended to unsuitable
The user of product sends the PUSH message of lipuid goods, it is possible to reduce the rubbish message that user receives prevents user due to excessive
Rubbish message generates dislike, promotes the likability of user.Further, since the Platform Server in store can regularly update store
Each user purchaser record and scoring record, it is ensured that selection be suitble to recommend this lipuid goods user when foundation reality
Shi Xingyu accuracys.
The step of various methods divide above, be intended merely to describe it is clear, when realization can be merged into a step or
Certain steps are split, multiple steps are decomposed into, as long as including identical logical relation, all in the protection domain of this patent
It is interior;To either adding inessential modification in algorithm in flow or introducing inessential design, but its algorithm is not changed
Core design with flow is all in the protection domain of the patent.
Four embodiment of the invention is related to a kind of device for recommending the commodity, as shown in figure 4, including:Logging modle 41, with note
It records the connected acquisition module 42 of module 41, the judgment module 43 being connected with acquisition module 42 and is pushed away with what judgment module 43 was connected
Send module 44, wherein
Logging modle 41 is mainly used for regularly updating the purchaser record of each user in store and scoring records;
Specifically, logging modle 41 includes inquiry submodule 411, sorting sub-module 412, and update submodule 413;
Inquiry submodule 411 is mainly used for periodically inquiring the commodity of the purchased item of each user with the first predetermined period
The scoring of number and purchased item records;
Sorting sub-module 412 is mainly used for after the inquiry each time of inquiry submodule, according to the purchased item of user
Quantity purchase, to purchased item carry out quantity sequence, and according to the scoring of purchased item record, to purchased item into
Row marking and queuing;Wherein, the more purchased item of quantity purchase, the ranking in quantity sequence are more forward;Scoring is higher
Purchased item, the ranking in marking and queuing are more forward;
Update submodule 413 is mainly used for the second predetermined period, is periodically sorted according to the quantity of each user and scoring is arranged
Sequence, the purchaser record and scoring for updating each user record;Wherein, the purchaser record of user include user quantity sequence in before
The goods number of X purchased item, the scoring of user record the preceding Y purchased item in the marking and queuing for including user
Goods number;X and Y is the natural number more than or equal to 1.
It is understood that the first predetermined period may be the same or different with the second predetermined period, the specific time
It can be chosen according to actual needs.Wherein, X and Y is the natural number more than or equal to 1;It is understood that X and Y can
To be chosen according to actual needs, value is bigger, then judges that the probability of push is bigger, value is smaller, then the user pushed gets over
Precisely.
Acquisition module 42 is mainly used for when needing to send the PUSH message of lipuid goods to user, from logging modle 41
The purchaser record and scoring for obtaining user record;
Judgment module 43 is mainly used for the purchaser record of the user obtained according to acquisition module 42 and scoring records, and judgement is
The no PUSH message that lipuid goods are sent to user;
Specifically, judgment module 43 includes the first detection sub-module 431 and the first decision sub-module 432;
First detection sub-module 431 is mainly for detection of in the purchaser record or scoring record of user, if including activity
The goods number of commodity;
First decision sub-module 432 is mainly used for detecting the goods number including lipuid goods in the first detection sub-module
When, sentence the PUSH message that directional user sends lipuid goods.
Pushing module 44 is used to, when judgment module 43 sentences directional user and sends the PUSH message of lipuid goods, send out to user
Send PUSH message.
Further, the scoring record of user further includes the goods number of the rear Z purchased item in marking and queuing;
Judgment module 43 further includes the second detection sub-module 433, third detection sub-module 434 and second judgment submodule
435;
Second detection sub-module 433 is mainly for detection of in the purchaser record of user, if the commodity including lipuid goods
Number;
Third detection sub-module 434 is mainly used for detecting the commodity including lipuid goods in the second detection sub-module 433
When number, the goods number of lipuid goods is detected, if be present in the goods number of Z purchased item;
Second decision sub-module 435 is mainly used for detecting the goods number of lipuid goods in third detection sub-module 434,
When being present in the goods number of Z purchased item, judgement does not send the PUSH message of lipuid goods to user;It is examined in third
Survey submodule and detect the goods numbers of lipuid goods, when being not present in the goods number of Z purchased item, judge to
Family sends the PUSH message of lipuid goods.
It is not difficult to find that present embodiment is device embodiment corresponding with first embodiment, present embodiment can be with
First embodiment is worked in coordination implementation.The relevant technical details mentioned in first embodiment still have in the present embodiment
Effect, in order to reduce repetition, which is not described herein again.Correspondingly, the relevant technical details mentioned in present embodiment are also applicable in
In first embodiment.
Fifth embodiment of the invention is related to a kind of device for recommending the commodity, and the 5th embodiment and the 4th embodiment are substantially
It is identical, the difference is that, further include anticipation module 45, as shown in figure 5,
Anticipation module 45 is mainly used for when needing to send the PUSH message of lipuid goods to user, and detection user is default
In the goods number for the commodity bought in duration, if there are the goods numbers of lipuid goods, and are detecting that there is no live
When the goods number of dynamic commodity, judgment module 43 is triggered;In the goods number for detecting the presence of lipuid goods, forbid to user
Send PUSH message.
Due in present embodiment other modules and its connection type in the fourth embodiment and act on all same,
Details are not described herein.
It is not difficult to find that present embodiment is device embodiment corresponding with second embodiment, present embodiment can be with
Second embodiment is worked in coordination implementation.The relevant technical details mentioned in second embodiment still have in the present embodiment
Effect, in order to reduce repetition, which is not described herein again.Correspondingly, the relevant technical details mentioned in present embodiment are also applicable in
In second embodiment.
It is noted that each module involved in present embodiment is logic module, and in practical applications, one
A logic unit can be a physical unit, can also be a part for a physical unit, can also be with multiple physics lists
The combination of member is realized.In addition, in order to protrude the innovative part of the present invention, it will not be with solution institute of the present invention in present embodiment
The technical issues of proposition, the less close unit of relationship introduced, but this does not indicate that there is no other single in present embodiment
Member.
It will be understood by those skilled in the art that the respective embodiments described above are to realize specific embodiments of the present invention,
And in practical applications, can to it, various changes can be made in the form and details, without departing from the spirit and scope of the present invention.
Claims (10)
1. a kind of Method of Commodity Recommendation, which is characterized in that it is applied to the Platform Server in store, including:
The purchaser record and scoring for regularly updating each user in the store record;
Detecting the purchaser record that each user is obtained when the lipuid goods information promoted and scoring record;
It is recorded according to the purchaser record of each user and scoring, judges whether to send pushing away for the lipuid goods to the user
Message is sent, and the PUSH message is sent when judgement sends the PUSH message to the user, then to the user.
2. Method of Commodity Recommendation according to claim 1, which is characterized in that each user for regularly updating the store
Purchaser record and scoring record, specifically include:
With the first predetermined period, the goods number of the purchased item of each user and the scoring note of purchased item are periodically inquired
Record;
After the inquiry each time, according to the quantity purchase of the purchased item of user, quantity row is carried out to purchased item
Sequence, and recorded according to the scoring of the purchased item, marking and queuing is carried out to purchased item, wherein quantity purchase is more
Purchased item, the ranking in quantity sequence is more forward, the higher purchased item of scoring, in the marking and queuing
In ranking it is more forward;
With the second predetermined period, periodically according to the sequence of the quantity of each user and the marking and queuing, the purchase of each user is updated
Buy record and scoring record, wherein the purchaser record of the user includes the preceding X in the quantity sequence of the user a
The goods number of commodity is bought, preceding Y that the scoring of the user records in the marking and queuing for including the user have purchased
The goods number of commodity is bought, the X and the Y are the natural number more than or equal to 1.
3. Method of Commodity Recommendation according to claim 2, which is characterized in that the purchaser record according to each user
It is recorded with scoring, judges whether the PUSH message for sending the lipuid goods to the user, specifically include:
In the purchaser record or the scoring record that detect the user, if include the goods number of the lipuid goods;
If including the goods number of the lipuid goods, the PUSH message that the lipuid goods are sent to the user is judged.
4. Method of Commodity Recommendation according to claim 2, which is characterized in that the purchaser record according to each user
It is recorded with scoring, judges whether before the PUSH message that the user sends the lipuid goods, further include:
In the goods number for detecting the commodity that the user has bought in preset duration, if there are the quotient of the lipuid goods
Product are numbered;
If there are the goods number of the lipuid goods, forbid sending the PUSH message to the user;
If there is no the goods numbers of the lipuid goods, then execute the purchaser record according to the user and scoring note
Record, judges whether the PUSH message that the lipuid goods are sent to the user.
5. Method of Commodity Recommendation according to claim 2, which is characterized in that the scoring of the user, which records, further includes:
The goods number of rear Z purchased item in the marking and queuing of the user, the Z are more than or equal to 1
Natural number;
The purchaser record according to the user and scoring record, and judge whether to send the lipuid goods to the user
PUSH message specifically includes:
In the purchaser record for detecting the user, if include the goods number of the lipuid goods;
If including the goods number of the lipuid goods, the goods number of the lipuid goods is detected, if be present in institute
In the goods number for stating Z purchased item;
If the goods number of the lipuid goods, it is present in the goods number of the Z purchased item, then judges not to institute
State the PUSH message that user sends the lipuid goods;If the goods number of the lipuid goods, it is not present in the Z and has purchased
It buys in the goods number of commodity, then judges the PUSH message for sending the lipuid goods to the user.
6. a kind of device for recommending the commodity, which is characterized in that including:
Logging modle, the purchaser record of each user for regularly updating store and scoring record;
Acquisition module obtains each user's for detecting when the lipuid goods information promoted from the logging modle
The purchaser record and scoring record;
Judgment module, the purchaser record of the user for being obtained according to the acquisition module and scoring record, judge
Whether the PUSH message of the lipuid goods is sent to the user;
Pushing module, for when the judgment module judges to send the PUSH message of the lipuid goods to the user, to
The user sends the PUSH message.
7. the device for recommending the commodity according to claim 6, which is characterized in that the logging modle specifically includes:
Submodule is inquired, for the first predetermined period, periodically inquiring the goods number of the purchased item of each user and having purchased
Buy the scoring record of commodity;
Sorting sub-module is used for after the inquiry described each time of the inquiry submodule, according to the purchased item of user
Quantity purchase carries out quantity sequence to purchased item, and is recorded according to the scoring of the purchased item, to purchased item
Carry out marking and queuing;Wherein, the more purchased item of quantity purchase, the ranking in quantity sequence are more forward;Scoring
Higher purchased item, the ranking in the marking and queuing are more forward;
Submodule is updated, for the second predetermined period, periodically being sorted and the marking and queuing according to the quantity of each user,
The purchaser record and scoring for updating each user record;Wherein, the purchaser record of the user includes the quantity of the user
The scoring record of the goods number of preceding X purchased item in sequence, the user includes the scoring row of the user
The goods number of preceding Y purchased item in sequence;The X and Y is the natural number more than or equal to 1.
8. the device for recommending the commodity according to claim 7, which is characterized in that the judgment module specifically includes:
First detection sub-module, during purchaser record or scoring for detecting the user record, if including the movable quotient
The goods number of product;
First decision sub-module, for detecting the goods number for including the lipuid goods in first detection sub-module
When, judge the PUSH message that the lipuid goods are sent to the user.
9. the device for recommending the commodity according to claim 7, which is characterized in that further include:
Anticipation module detects the user and has been purchased in preset duration for detecting when the lipuid goods information promoted
In the goods number for the commodity bought, if there are the goods numbers of the lipuid goods, and the activity is not present detecting
When the goods number of commodity, the judgment module is triggered;When detecting the presence of the goods number of the lipuid goods, forbid to
The user sends the PUSH message.
10. the device for recommending the commodity according to claim 7, which is characterized in that the scoring of the user, which records, further includes:
The goods number of rear Z purchased item in the marking and queuing of the user, the Z are more than or equal to 1
Natural number;
The judgment module specifically includes:
Second detection sub-module, in the purchaser record for detecting the user, if the commodity including the lipuid goods are compiled
Number;
Third detection sub-module, for detecting the goods number for including the lipuid goods in second detection sub-module
When, detect the goods number of the lipuid goods, if be present in the goods number of the Z purchased item;
Second decision sub-module, the goods number for detecting the lipuid goods in the third detection sub-module, exists
When in the goods number of the Z purchased item, judgement does not send the PUSH message of the lipuid goods to the user;
The goods number of the lipuid goods is detected in the third detection sub-module, is not present in the Z purchased item
When in goods number, the PUSH message that the lipuid goods are sent to the user is judged.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810177406.5A CN108346090A (en) | 2018-02-28 | 2018-02-28 | Method of Commodity Recommendation and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810177406.5A CN108346090A (en) | 2018-02-28 | 2018-02-28 | Method of Commodity Recommendation and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108346090A true CN108346090A (en) | 2018-07-31 |
Family
ID=62959021
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810177406.5A Pending CN108346090A (en) | 2018-02-28 | 2018-02-28 | Method of Commodity Recommendation and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108346090A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112381623A (en) * | 2020-12-04 | 2021-02-19 | 南京浪脆电子商务有限公司 | Electronic commerce platform commodity intelligent recommendation method based on big data analysis |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345695A (en) * | 2013-06-25 | 2013-10-09 | 百度在线网络技术(北京)有限公司 | Commodity recommendation method and device |
CN106485562A (en) * | 2015-09-01 | 2017-03-08 | 苏宁云商集团股份有限公司 | A kind of commodity information recommendation method based on user's history behavior and system |
CN106600302A (en) * | 2015-10-19 | 2017-04-26 | 玺阅信息科技(上海)有限公司 | Hadoop-based commodity recommendation system |
CN107292645A (en) * | 2016-03-31 | 2017-10-24 | 壹贰叁叁购(厦门)信息技术有限公司 | A kind of Method of Commodity Recommendation of dynamic self-adapting |
CN107679898A (en) * | 2017-09-26 | 2018-02-09 | 浪潮软件股份有限公司 | A kind of Method of Commodity Recommendation and device |
-
2018
- 2018-02-28 CN CN201810177406.5A patent/CN108346090A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345695A (en) * | 2013-06-25 | 2013-10-09 | 百度在线网络技术(北京)有限公司 | Commodity recommendation method and device |
CN106485562A (en) * | 2015-09-01 | 2017-03-08 | 苏宁云商集团股份有限公司 | A kind of commodity information recommendation method based on user's history behavior and system |
CN106600302A (en) * | 2015-10-19 | 2017-04-26 | 玺阅信息科技(上海)有限公司 | Hadoop-based commodity recommendation system |
CN107292645A (en) * | 2016-03-31 | 2017-10-24 | 壹贰叁叁购(厦门)信息技术有限公司 | A kind of Method of Commodity Recommendation of dynamic self-adapting |
CN107679898A (en) * | 2017-09-26 | 2018-02-09 | 浪潮软件股份有限公司 | A kind of Method of Commodity Recommendation and device |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112381623A (en) * | 2020-12-04 | 2021-02-19 | 南京浪脆电子商务有限公司 | Electronic commerce platform commodity intelligent recommendation method based on big data analysis |
CN112381623B (en) * | 2020-12-04 | 2021-07-06 | 深圳市快云科技有限公司 | Electronic commerce platform commodity intelligent recommendation method based on big data analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Holt | Brands and branding | |
Leskovec et al. | The dynamics of viral marketing | |
CN103841122B (en) | Target object information recommends method, server and client | |
US9542503B2 (en) | Estimation of closeness of topics based on graph analytics | |
Angwin | The web’s new gold mine: Your secrets | |
US8438062B2 (en) | Network node ad targeting | |
CN109325182B (en) | Information pushing method and device based on session, computer equipment and storage medium | |
CN101589360A (en) | Generic online ranking system that is suitable for uniting and method | |
KR100800014B1 (en) | Multi-level marketing management system and business method using the same | |
CN110799980B (en) | Pictographic understanding in online experience | |
CN104620276A (en) | Attendee suggestion for events based on profile information on a social networking site | |
CN101438279A (en) | Search system and methods with integration of user annotations from a trust network | |
US20100250366A1 (en) | Merge real-world and virtual markers | |
CN105913301A (en) | Bill sharing information processing method and system | |
US20230103440A1 (en) | Ranking data objects using machine learning and scoring to configure a graphical interface | |
WO2014139372A1 (en) | System and method for selecting and rendering content | |
JP2018045553A (en) | Selection device, selection method, and selection program | |
KR100744063B1 (en) | Internet search advertising service system and method thereof | |
US20140108152A1 (en) | Managing Social Network Relationships Between A Commercial Entity and One or More Users | |
CN107070932A (en) | The anonymous methods for preventing label neighbours from attacking in community network dynamic publishing | |
CN108346090A (en) | Method of Commodity Recommendation and device | |
KR100754855B1 (en) | Internet keyword advertising method using secondary keyword, and Internet keyword advertising system thereof | |
CN108846034A (en) | A method of about user behavior analysis | |
CN105871700A (en) | Message propagation method and server | |
Zhao et al. | The marketing effects of recommender systems in a B2C e-commerce context: A review and future directions |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180731 |
|
RJ01 | Rejection of invention patent application after publication |