CN106484747A - A kind of webpage item recommendation method based on alternative events and device - Google Patents
A kind of webpage item recommendation method based on alternative events and device Download PDFInfo
- Publication number
- CN106484747A CN106484747A CN201510557884.5A CN201510557884A CN106484747A CN 106484747 A CN106484747 A CN 106484747A CN 201510557884 A CN201510557884 A CN 201510557884A CN 106484747 A CN106484747 A CN 106484747A
- Authority
- CN
- China
- Prior art keywords
- alternative events
- event
- degree
- interest
- score value
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Transfer Between Computers (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the present application provides a kind of webpage item recommendation method based on alternative events and device.Methods described includes:According to multiple users for the interactive information of the alternative events of webpage project, the degree of correlation of alternative events and the targeted interactive event in alternative events is generated;For the alternative events of active user, the corresponding target web project of at least one targeted interactive event is extracted according to alternative events and the degree of correlation of multiple targeted interactive event;By target web project recommendation to active user, it is to avoid lead to not carry out the problem of effective webpage project recommendation because user does not provide evaluation point;And, using the alternative events for webpage project as the basis of relatedness computation, the webpage project recommendation of user interest can will be met objective, exactly to user;Further, the degree of correlation of targeted interactive event and alternative events is only calculated in relatedness computation, is reduced the data volume in relatedness computation, is saved computing resource and the storage resource of server.
Description
Technical field
The application is related to Internet technical field, more particularly to a kind of webpage project based on alternative events
Recommendation method and a kind of webpage project recommendation device based on alternative events.
Background technology
With the continuous development of Internet technology, increasing user browses extensive stock on webpage.
For webpage project that may be interested to user's recommended users, websites providers can be collected in a large number
User divides to the evaluation of each commodity, and is divided by collaborative filtering Utilization assessment and calculate between commodity
Correlation matrix.For certain user, once the evaluation that part commodity are made was divided using the user,
And the part commodity and the degree of correlation of other commodity, calculate the recommendation degree of other commodity and correspondingly push away
Recommend to the user.
There are three problems in this Method of Commodity Recommendation at present:First, in actual applications, user may
The evaluation for being not desired to spend time and efforts to provide commodity divides, and causes in default of evaluation point and cannot be had
The commercial product recommending of effect;Secondly, user divides with subjective tendency for the evaluation of commodity, is commented based on subjective
Valency divides carries out commercial product recommending, it is impossible to objective, will meet the commercial product recommending of user interest exactly to user;
Finally, as the data volume of the commodity in internet is on the increase, commodity are carried out by collaborative filtering
Recommend, the data volume of the correlation matrix in current algorithm can be caused drastically to expand, cause server to overstock
Deposit, process work excess load is calculated, or even cause server crash.
Content of the invention
In view of the above problems, it is proposed that the embodiment of the present application overcomes the problems referred to above or extremely to provide one kind
A kind of webpage item recommendation method based on alternative events for partially solving the above problems is with accordingly
A kind of webpage project recommendation device based on alternative events.
In order to solve the above problems, this application discloses a kind of webpage project recommendation side based on alternative events
Method, including:
According to multiple users for the interactive information of the alternative events of webpage project, the alternative events are generated
The degree of correlation with the targeted interactive event in the alternative events;
For the alternative events of active user, according to the phase of the alternative events and multiple targeted interactive event
Guan Du extracts the corresponding target web project of at least one targeted interactive event;
The active user is given in the target web project recommendation.
Preferably, described according to multiple users for the interactive information of the alternative events of webpage project, generate
The step of degree of correlation of the targeted interactive event in the alternative events and the alternative events, includes:
The interest-degree score value of the alternative events is generated according to the event attribute of the alternative events;
The alternative events are calculated according to the interest-degree score value of the alternative events thing is interacted with the target
The degree of correlation of part.
Preferably, the alternative events include the browsing event to webpage project, and/or collection event and
/ or add shopping cart event, and/or purchase events, described according to the event attribute of alternative events generation institute
The step of interest-degree score value for stating alternative events, includes:
With the browsing event, the collection event, the interpolation shopping cart event, the purchase events
In alternative events execute the interest-degree of the product as the then alternative events of times N and preset parameter
Score value is M.
Preferably, described according to the interest-degree score value of the alternative events calculating alternative events and the target
The step of degree of correlation of alternative events, includes:
The alternative events of same user are constituted event pair with targeted interactive event, and calculates the event pair
In alternative events and targeted interactive event interest-degree score value product;
Interest-degree score value according to the alternative events of the identical event centering of multiple users, target interaction thing
The interest-degree score value of part and alternative events and the product of the interest-degree score value of targeted interactive event, calculate
The alternative events of the event centering and the degree of correlation for handing over targeted interactive event.
Preferably, the interest-degree of the alternative events of the identical event centering according to multiple users divides
The interest-degree of value, the interest-degree score value of targeted interactive event and alternative events and targeted interactive event divides
The product of value, the alternative events for calculating the event centering are concrete with the degree of correlation for handing over targeted interactive event
For:
For the identical event pair of multiple users, the interest of the alternative events of identical event centering is calculated
Square sum, square sum of the interest-degree score value of targeted interactive event of degree score value and alternative events
With the sum of products of the interest-degree score value of targeted interactive event, and the interaction of the event centering is correspondingly formed
Event multi-C vector value corresponding with targeted interactive event difference;
Calculate cosine value between the multi-C vector value and as the degree of correlation.
Preferably, described according to the interest-degree score value of the alternative events calculate the alternative events with described
The step of degree of correlation of targeted interactive event, includes:
The alternative events of same user are constituted event pair with targeted interactive event, and using multiple users'
The alternative events of identical event centering are separately constituted with the interest-degree score value of targeted interactive event and interact thing
The set of part interest-degree score value and targeted interactive event interest-degree score value set;
Alternative events interest-degree score value set is calculated with the targeted interactive event interest-degree score value collection
The outstanding person of conjunction blocks German number and as the degree of correlation.
Preferably, when the active user there occurs alternative events, described for active user's
Alternative events, extract at least one mesh according to the degree of correlation of the alternative events with multiple targeted interactive event
The step of mark alternative events corresponding target web project, includes:
When alternative events for capturing user generation, search one alternative events with multiple
The degree of correlation of targeted interactive event;
Extract the degree of correlation and meet the corresponding webpage project conduct of targeted interactive event in the first preset range
Target web project.
Preferably, when the active user there occurs multiple alternative events, described for active user's
Alternative events, extract at least one mesh according to the degree of correlation of the alternative events with multiple targeted interactive event
The step of mark alternative events corresponding target web project, includes:
For each targeted interactive event, search the targeted interactive event respectively with multiple alternative events
The degree of correlation;
According to the targeted interactive event degree of correlation respectively with each alternative events, the mark mesh is calculated
Mark the recommendation degree of alternative events and the correlation of multiple alternative events;
Extract recommendation degree and meet the corresponding webpage project conduct of targeted interactive event in the second preset range
Target web project.
Preferably, the webpage project includes trading object, and/or video, and/or audio frequency, and/or electricity
Sub- reading matter.
Disclosed herein as well is a kind of webpage project recommendation device based on alternative events, including:
Degree of correlation generation module, for believing for the interaction of the alternative events of webpage project according to multiple users
Breath, generates the degree of correlation of the targeted interactive event in the alternative events and the alternative events;
Extraction module, for the alternative events for active user, according to the alternative events and multiple mesh
The degree of correlation of mark alternative events extracts the corresponding target web project of at least one targeted interactive event;
Recommending module, for giving the active user by the target web project recommendation.
Preferably, the degree of correlation generation module includes:
Interest-degree score value generates submodule, for generating the friendship according to the event attribute of the alternative events
The interest-degree score value of mutual event;
Relatedness computation submodule, for calculating the interaction according to the interest-degree score value of the alternative events
The degree of correlation of event and the targeted interactive event.
Preferably, the alternative events include the browsing event to webpage project, and/or collection event and
/ or adding shopping cart event, and/or purchase events, the interest-degree score value generates submodule to be included:
Interest-degree score value M generates subelement, for the browsing event, the collection event, institute
State shopping cart event of adding, the alternative events in the purchase events and execute taking advantage of for times N and preset parameter
Product is M as the interest-degree score value of the then alternative events.
Preferably, the relatedness computation submodule includes:
Event to constitute subelement, for the alternative events of same user are succeeded with targeted interactive event group
Part pair, and calculate taking advantage of for the alternative events of the event centering and the interest-degree score value of targeted interactive event
Product;
Relatedness computation subelement, for according to the alternative events of the identical event centering of multiple users
Interest-degree score value, the interest-degree score value of targeted interactive event and alternative events and targeted interactive event
The product of interest-degree score value, the alternative events for calculating the event centering are related to hand over targeted interactive event
Degree.
Preferably, the relatedness computation subelement specifically for:
For the identical event pair of multiple users, the interest of the alternative events of identical event centering is calculated
Square sum, square sum of the interest-degree score value of targeted interactive event of degree score value and alternative events
With the sum of products of the interest-degree score value of targeted interactive event, and the interaction of the event centering is correspondingly formed
Event multi-C vector value corresponding with targeted interactive event difference;
Calculate cosine value between the multi-C vector value and as the degree of correlation.
Preferably, the relatedness computation submodule includes:
Interest-degree score value collection is combined into subelement, for the alternative events of same user are interacted thing with target
Part constitutes event pair, and interacts thing using the alternative events of the identical event centering of multiple users with target
The interest-degree score value of part separately constitutes the set of alternative events interest-degree score value and targeted interactive event interest-degree
Score value set;
The German several computation subunit of outstanding card, for calculate alternative events interest-degree score value set with described
The outstanding person of targeted interactive event interest-degree score value set blocks German number and as the degree of correlation.
Preferably, when the active user there occurs alternative events, the extraction module includes:
First degree of correlation searches submodule, for working as alternative events for capturing user generation,
Search the degree of correlation of one alternative events and multiple targeted interactive event;
First extracting sub-module, meets the targeted interactive event in the first preset range for extracting the degree of correlation
Corresponding webpage project is used as target web project.
Preferably, when the active user there occurs multiple alternative events, the extraction module includes:
Second degree of correlation searches submodule, for being directed to each targeted interactive event, searches the target and hands over
The mutual event degree of correlation respectively with multiple alternative events;
Recommendation degree calculating sub module, for according to the targeted interactive event respectively with each alternative events
The degree of correlation, calculates the recommendation degree of the mark targeted interactive event and the correlation of multiple alternative events;
Second extracting sub-module, meets the targeted interactive event in the second preset range for extracting recommendation degree
Corresponding webpage project is used as target web project.
Preferably, the webpage project includes trading object, and/or video, and/or audio frequency, and/or electricity
Sub- reading matter.
The embodiment of the present application includes advantages below:
The embodiment of the present application carries out relatedness computation based on user to the alternative events of webpage project, keeps away
Exempt from because user does not provide evaluation point and led to not carry out asking for effective webpage project recommendation
Topic;And, using for webpage project alternative events as relatedness computation basis, permissible
Objective, the webpage project recommendation of user interest will be met exactly to user;Further, in correlation
Degree only calculates the degree of correlation of targeted interactive event and alternative events in calculating, reduce relatedness computation
In data volume, save computing resource and the storage resource of server.
Description of the drawings
The step of Fig. 1 is a kind of webpage item recommendation method embodiment 1 based on alternative events of the application
Flow chart;
The step of Fig. 2 is a kind of webpage item recommendation method embodiment 2 based on alternative events of the application
Flow chart;
Fig. 3 is a kind of structural frames of webpage project recommendation device embodiment based on alternative events of the application
Figure;
Fig. 4 is that the embodiment of the present application 2 calculates targeted interactive event and alternative events using Map Reduce
The degree of correlation flow chart.
Specific embodiment
Understandable for enabling the above-mentioned purpose of the application, feature and advantage to become apparent from, below in conjunction with the accompanying drawings
The application is described in further detail with specific embodiment.
With reference to Fig. 1, show that a kind of webpage item recommendation method based on alternative events of the application is implemented
The step of example 1 flow chart, specifically may include steps of:
Step 101, according to multiple users for the interactive information of the alternative events of webpage project, generates institute
State the degree of correlation of the targeted interactive event in alternative events and the alternative events.
It should be noted that above-mentioned webpage project can include webpage in various trading objects, video,
Audio frequency or electron reading etc..User can get corresponding information or clothes by different webpage projects
Business.
In practical application scene, user can be directed to different webpage projects according to oneself degree interested
Produce different alternative events.For example, user is interested in certain trading object in webpage, can be to this
The alternative events browsed by trading object, may be after multiple browsing, and user is to the trading object
Level of interest lifting, and produce one by trading object collect alternative events or be positioned over webpage
The alternative events of the virtual shopping cart of offer, end user are quite interested in the trading object, or even meeting
The alternative events that is bought.From above-mentioned actual scene, user is for same webpage project
Distinct interaction event can reflect user's degree interested in the webpage project, therefore, it can net
What page project and user was directed to that the interbehavior of webpage project is combined together to form webpage project interacts thing
Part.
Meanwhile, can also there is correlation between each alternative events of different webpage projects in user.
For example, the alternative events browsed to trading object A by same user, are carried out to trading object B
The alternative events that buys, it can be understood as, there are certain interest the use for being browsed to trading object A
Family, it is also possible to produce considerably more intense interest, and the interaction that is finally bought to trading object B
Event.Can be seen that with implicit correlation between the alternative events of webpage project from the example, this
Plant implicit correlation and can reflect that user is related to the level of interest of different webpage projects.
User can be reflected to the alternative events of webpage project using the interactive information of alternative events
Interest level be quantized into specific interest-degree score value M.Above-mentioned interactive information can include interaction thing
The alternative events classification of part, alternative events execute number of times, alternative events time of origin and alternative events and continue
At least one among time etc..For example, user is 3 times to the number of visits of webpage project A, then net
The interest-degree score value M of the browsing event of page project A is 3.Certainly, those skilled in the art can root
Weight is arranged to different alternative events according to being actually needed, to distinguish the importance of different alternative events.
The correlation between alternative events can be quantized into the degree of correlation using specific interest-degree score value.Tool
During gymnastics is made, can be respectively constituted using the interest-degree score value of two alternative events of identical of a large number of users
The multi-C vector of two interest-degree score values, calculates the cosine between two multi-C vectors using cosine value formula
Value, the cosine value are the degree of correlation between two alternative events.
Certainly, those skilled in the art can also calculate alternative events using other modes according to actual needs
Between the degree of correlation, for example, if certain user once executed certain alternative events, interest-degree divides
Value can be defined as 1, if be not carried out, interest-degree score value can be defined as 0, using a large amount of
The interest-degree score value of two alternative events of identical of user respectively constitutes the multidimensional of two interest-degree score values
Vector, and the German number of the card of the outstanding person between two multi-C vectors is calculated, the outstanding person blocks German number and is two interactions
The degree of correlation between thing.
However, the degree of correlation between not all of alternative events all merits attention and calculates, because user
In alternative events to webpage project, only part alternative events represent user which is had strong emerging
Interest, for example, it is the strongest to the level of interest of the webpage project that the alternative events of purchase represent user,
And the alternative events for browsing may be user simply understands, the webpage project for browsing simultaneously does not meet user
Interest.The purpose of webpage project recommendation be in order to most possibly be caused user interest, be best suited user
The webpage project recommendation of interest is to user, therefore it may only be necessary to the targeted interactive event in alternative events is paid close attention to,
Calculate the degree of correlation of targeted interactive event and other alternative events.In concrete implementation, can adopt
With the interactive information of alternative events, the targeted interactive event in alternative events and non-targeted interaction thing is distinguished
Part, calculates the degree of correlation of targeted interactive event and other alternative events, so as to obtain the target of webpage project
The degree of correlation of targeted interactive event or non-targeted alternative events of the event with another webpage project.As this
Bright preferred exemplary, can be stored in the degree of correlation for calculating in database.
Those skilled in the art can be according to actual needs using one or more interaction in alternative events
Event is not restricted to this as targeted interactive event, the embodiment of the present invention.
Step 102, for the alternative events of active user, is handed over multiple targets according to the alternative events
Mutually the degree of correlation of event extracts the corresponding target web project of at least one targeted interactive event.
Step 103, the active user is given in the target web project recommendation.
In actual application, alternative events of the active user to certain webpage project, with multiple and its
The degree of correlation of the targeted interactive event of his webpage project, can be by the one or more phases in preset range
The corresponding target web item extraction of the targeted interactive event of Guan Du simultaneously recommends active user.
Divide the degree of correlation between calculating webpage project compared with current to the evaluation of webpage project based on user
Method, webpage project is combined together to form and interacts with the interbehavior for webpage project by the application
Event, carry out relatedness computation based on alternative events, it is to avoid leads because user does not provide evaluation point
Cause to carry out the problem of effective webpage project recommendation;And, using the interactive thing for webpage project
The webpage project for meeting user interest can be pushed away objective, exactly by part as the basis of relatedness computation
Recommend to user;Further, it is related to alternative events that targeted interactive event is only calculated in relatedness computation
Degree, reduces the data volume in relatedness computation, saves computing resource and the storage resource of server.
With reference to Fig. 2, show that a kind of webpage item recommendation method based on alternative events of the application is implemented
The step of example 2 flow chart, specifically may include steps of:
Step 201, generates the interest-degree score value of the alternative events according to the event attribute of alternative events.
Above-mentioned event attribute can include that the alternative events of alternative events execute number of times, alternative events and occur
At least one among time and alternative events duration etc..Belonged to according to the various events of alternative events
Property, a specific interest-degree score value of alternative events can be generated.In actual applications, when user not
There are certain alternative events, then the interest-degree score value of the alternative events of user can be 0.
Used as the preferred exemplary of the embodiment of the present invention, the alternative events include to browse thing to webpage project
Part, and/or collection event, and/or interpolation shopping cart event, and/or purchase events, the step 201
Following sub-step can be included:
Sub-step S11, with the browsing event, the collection event, the interpolation shopping cart event,
Alternative events in the purchase events are executed times N and are interacted as then described with the product of preset parameter
The interest-degree score value of event is M.
User generally includes to browse, collects, adds shopping cart, purchase etc. to the alternative events of webpage project
Alternative events.On the one hand, user generally reflects to webpage project to the execution number of times of certain alternative events
Level of interest.On the other hand, different alternative events also have different significance levels, for example, buy
Event is of paramount importance event, adds shopping cart and takes second place, browses, be the usual event of comparison, for
The significance level of event gives preset parameter as the weight of event, and interest-degree score value can be caused more reasonable
Ground reflection level of interest of the user to webpage project.
Step 202, calculates the alternative events and the mesh according to the interest-degree score value of the alternative events
The degree of correlation of mark alternative events.
Used as the preferred exemplary one of the embodiment of the present invention, the step 202 can include following sub-step:
Sub-step S21, the alternative events of same user is constituted event pair with targeted interactive event, and is counted
Calculate the product of the alternative events of the event centering and the interest-degree score value of targeted interactive event.
The alternative events of same user can be constituted event pair with targeted interactive event, by event centering
The interest-degree score value of alternative events is multiplied with the interest-degree score value of targeted interactive event.Hand over when there is no target
Mutually during event, i.e., the interest-degree score value of targeted interactive event is 0, then the result being multiplied also is 0.Because
Only need to pay close attention to the degree of correlation of targeted interactive event and alternative events, when there is no targeted interactive event,
The event can save follow-up relatedness computation to the multiplied result for 0.
Sub-step S22, interest-degree score value according to the alternative events of the identical event centering of multiple users,
The interest-degree score value of targeted interactive event and alternative events and the interest-degree score value of targeted interactive event
Product, calculates the alternative events of the event centering and the degree of correlation of targeted interactive event.
Alternative events and the degree of correlation of targeted interactive event, can be the cosine values of both interest-degree score values,
Can also be that the outstanding person of both interest-degree score values blocks German number.Those skilled in the art according to circumstances can adopt
Any similarity degree that can be calculated between alternative events and the interest-degree score value of targeted interactive event
Algorithm calculates the degree of correlation.
Used as the preferred exemplary of the embodiment of the present invention, sub-step S22 can be specifically included:
Sub-step S22-1, for the identical event pair of multiple users, calculates identical event centering
Square sum of the interest-degree score value of alternative events, the interest-degree score value of targeted interactive event square it
With and alternative events and targeted interactive event interest-degree score value the sum of products, and be correspondingly formed institute
State the alternative events multi-C vector value corresponding with targeted interactive event difference of event centering;
Sub-step S22-2, calculates cosine value between the multi-C vector value and as the degree of correlation.
For the identical event pair of multiple users, multiple interest-degrees to same targeted interactive event divide
Value, and multiple interest-degree score values to same alternative events, form targeted interactive event and alternative events
The corresponding multi-C vector value of difference, calculates the cosine value between multi-C vector value using cosine value formula, with
This cosine value is used as the event centering targeted interactive event and the degree of correlation of alternative events.
In practical application, it is possible to use Map Reduce carries out the calculating of the degree of correlation with reference to cosine degree formula.
The flow process of targeted interactive event and the degree of correlation of alternative events is calculated with reference to the utilization Map Reduce of Fig. 4
Figure, realizes the calculating of the degree of correlation by 5 Map Reduce processes.
Cosine value formula can be:Wherein, SijFor targeted interactive event
The degree of correlation with alternative events.
Map Reduce 1, collects all alternative events of same user, and alternative events are handed over target
Mutually event forms event pair, calculates event centering alternative events with the interest-degree score value of targeted interactive event
Product.Wherein,<ei、ej>For the event pair that alternative events are formed with targeted interactive event;ruiFor interaction
The interest-degree score value of event;rujInterest-degree score value for targeted interactive event.
Map Reduce 2, calculates multiple user's identical event centering targeted interactive event and alternative events
Interest-degree score value the sum of products.Wherein,For multiple user's identical event centering targets
The sum of products of the interest-degree score value of alternative events and alternative events.
Map Reduce 3, calculates the interest-degree score value of multiple user's identical event centering alternative events
Quadratic sum.Wherein,Put down for the interest-degree score value of multiple user's identical event centering alternative events
Square sum.
Map Reduce 4, the interest-degree for calculating different user identical event centering targeted interactive event divide
The quadratic sum of value.Wherein,Interest for multiple user's identical event centering targeted interactive event
Degree score value square sum.
Map Reduce 5, calculates multiple user's identical event centering targeted interactive event and alternative events
Interest-degree score value cosine value.Wherein,Interact for different user's identical event centerings
The evolution of the interest-degree score value square sum of event;For different user's identical event centerings
The evolution of the interest-degree score value square sum of targeted interactive event.
Used as the preferred exemplary two of the embodiment of the present invention, the step 202 can include following sub-step:
Sub-step S31, the alternative events of same user is constituted event pair with targeted interactive event, and is adopted
Divided with the alternative events of identical event centering and the interest-degree score value of targeted interactive event of multiple users
Zu Cheng not the set of alternative events interest-degree score value and targeted interactive event interest-degree score value set.
Sub-step S32, calculates alternative events interest-degree score value set emerging with the targeted interactive event
The outstanding person of interesting degree score value set blocks German number and as the degree of correlation.
In actual application, when user does not produce certain alternative events to webpage project, then the interaction
The interest-degree score value of event can be 0, if there is certain alternative events, then the interest of the alternative events
Degree score value can be 1.Using multiple user's identical event centering alternative events with targeted interactive event it is
1 or 0 interest-degree score value, may be constructed and correspond respectively to all of alternative events and targeted interactive event
Dimension is 1 or 0 multi-C vector value, calculates dimension and is between 1 or 0 two multi-C vector values
The German number of outstanding card, and the outstanding person for calculating is blocked German number as the phase of alternative events and targeted interactive event
Guan Du.The degree of correlation is calculated using the German number of outstanding card to only need to pay close attention to presence or absence two states suitable for calculating
Alternative events between the degree of correlation.
Step 203, for the alternative events of active user, is handed over multiple targets according to the alternative events
Mutually the degree of correlation of event extracts the corresponding target web project of at least one targeted interactive event.
User may produce one or more alternative events to the webpage project in webpage, be produced according to user
Raw alternative events, can be by corresponding for the targeted interactive event related to alternative events target web item
Mesh is extracted.In practical application, it is possible that the quantity of the targeted interactive event of correlation is larger, and portion
The degree of correlation of the alternative events that partial objectives for alternative events are produced with user is relatively low, need not carry therefore and all
The corresponding webpage project of targeted interactive event is taken, can be by targeted interactive event according to the phase with alternative events
The size of Guan Du is ranked up, and extracts the corresponding webpage project of the forward targeted interactive event of sequence, also may be used
The degree of correlation, to preset the scope of a degree of correlation, is met the corresponding net of targeted interactive event of the scope
Page item extraction.Those skilled in the art can also extract target web project using other modes, for example,
By calculating the recommendation degree of targeted interactive event, when recommendation degree meets preset range, by the target interaction
The corresponding webpage project of event is as target web project recommendation to user.
As the preferred exemplary one of the embodiment of the present invention, when the active user there occurs alternative events
When, the step 203 can include following sub-step:
Sub-step S41, when alternative events for capturing user generation, searches one friendship
The mutual degree of correlation of event and multiple targeted interactive event.
Sub-step S42, extracts the degree of correlation and meets the corresponding net of targeted interactive event in the first preset range
Page project is used as target web project.
Alternative events are generated to certain webpage project when user is captured, with multiple and other webpage items
The degree of correlation of purpose targeted interactive event, can be by the mesh of the one or more degrees of correlation in preset range
The corresponding target web item extraction of mark alternative events, in order to follow-up webpage project recommendation.By pre-
If the corresponding webpage project of the targeted interactive event that scope can targetedly recommend the degree of correlation higher.
Preset range can be configured according to actual conditions by those skilled in the art, for example, can be available for
The quantity adjustment preset range of the corresponding webpage project of the targeted interactive event of extraction, when the mesh for being available for extracting
The quantity of the corresponding webpage project of mark alternative events is very few, can reduce the lower limit of preset range, it is ensured that have
Enough webpage projects can recommend active user.
In practical application, distributed file system (Hadoop Distributed File can be advanced with
System, Hadoop) large construction cluster calculates the degree of correlation of alternative events and targeted interactive event, works as seizure
Alternative events are generated to user for certain webpage project, can search and the friendship in database
Mutually the related targeted interactive event of event and its degree of correlation, extract the target friendship that the degree of correlation meets preset range
The mutual corresponding target web project of event.
As the preferred exemplary two of the embodiment of the present invention, when the active user there occurs multiple alternative events
When, the step 203 can include following sub-step:
Sub-step S51, for each targeted interactive event, search the targeted interactive event respectively with many
The degree of correlation of individual alternative events.
Sub-step S52, according to the targeted interactive event degree of correlation respectively with each alternative events, counts
Calculate the recommendation degree of the mark targeted interactive event and the correlation of multiple alternative events.
Sub-step S53, extracts recommendation degree and meets the corresponding net of targeted interactive event in the second preset range
Page project is used as target web project.
In practical application, user may produce multiple alternative events, the plurality of alternative events may with
There is correlation in one targeted interactive event, using the interest-degree score value of multiple alternative events and multiple
Alternative events and the degree of correlation of above-mentioned targeted interactive event, can calculate pushing away for the targeted interactive event
Corresponding for targeted interactive event webpage project, when recommendation degree meets default scope, can be made by degree of recommending
For target web project.By using the degree of correlation as weight calculation targeted interactive event recommendation degree, with
Recommendation degree is used as the basis of webpage project recommendation, it is recommended that more meet the emerging of user to the webpage project of user
Interest.
Step 204, the active user is given in the target web project recommendation.
Above-mentioned webpage project can include trading object, and/or video, and/or audio frequency, and/or electronics
Reading matter.
The embodiment of the present application is divided by generating the interest-degree of alternative events according to the event attribute of alternative events
User can be objectively quantized into interest-degree score value to the level of interest of webpage project, based on interest by value
Degree score value carries out webpage project recommendation, and the webpage project that is recommended more meets the interest of user.
And, the corresponding target web project recommendation of the targeted interactive event by the degree of correlation in preset range
To user, the flexibility of webpage project recommendation is improved.
For making those skilled in the art more fully understand the application, below by way of a specific example explanation
According to multiple users for the interactive information of the alternative events of webpage project in the application, alternative events are generated
Method with the degree of correlation of the targeted interactive event in the alternative events:
Have in the application scenarios of three users A, B and C to two webpage projects x and y at one,
Browsing event for webpage project x and webpage project y, collection event, placement shopping cart event, purchase
Formal matter part is respectively E1、E2、E3、E4、E5、E6、E7And E8.In this example, shopping is made
For targeted interactive event, therefore E4And E8For the targeted interactive event in the example.
May browse through event, collection event, the alternative events that adds in shopping cart event, purchase events to hold
Row times N is M with the product of preset parameter as the interest-degree score value of the then alternative events.
User A, B and C for the browsing event of webpage project x and y, collection event, place shopping
The execution number of times E of the alternative events such as car event, shopping event is respectively EAx is browsed=3, EAx is collected=2, EAx Place shopping cart=1, EAx does shopping=2, EAy is browsed=2, EAy is collected=1, EAy places shopping cart=1, EAy does shopping=0, EBx is browsed=5,
EBx is collected=1, EBx places shopping cart=1, EBx does shopping=1, EBy is browsed=2, EBy is collected=2, EBy places shopping cart=2, EBy does shopping
=1, ECx is browsed=2, ECx is collected=1, ECx places shopping cart=1, ECx does shopping=1, ECy is browsed=3, ECy is collected=2, ECy Place shopping cart=1 and ECy does shopping=1.
Browsing event, collection event, place shopping cart event, shopping event preset parameter be respectively 1,
2nd, 3,4, interacted as this using the number of times that executes of alternative events with the product of the preset parameter of alternative events
The interest-degree score value of event, from obtained from each alternative events interest-degree score value as shown in the table:
The alternative events of same user are constituted event pair with targeted interactive event, and calculates the event pair
In alternative events and targeted interactive event interest-degree score value product.For easy description, now only need
E to be calculated4And E5、E2And E8、E1And E4The degree of correlation, it is possible to use cosine degree formula calculate S14、
S54And S28, wherein, concrete calculating process is as follows:
Alternative events E by user A5With targeted interactive event E4Composition event pair<E5、E4>, event
Right<E5、E4>In, E5And E4Interest-degree score value product be rA5*rA4=2*8=16.Same calculating
Method can obtain the event pair of user A<E2、E8>In, E2And E8The product of interest-degree score value be
rA2*rA8=4*0=0;Also the event pair of user A is obtained<E1、E4>In, E1And E4Interest-degree score value
Product be rA1*rA4=3*8=24.
According to the method described above, the event pair of user B can be obtained<E5、E4>, event pair<E2、E8>With
Event pair<E1、E4>In, E5And E4Interest-degree score value product be 8, E2And E8Interest-degree divide
The product of value is 8, E1And E4Interest-degree score value product be 20.
According to the method described above, the event pair of user C can be obtained<E5、E4>, event pair<E2、E8>With
Event pair<E1、E4>In, E4And E5Interest-degree score value product be 12, E2And E8Interest-degree divide
The product of value is 8, E1And E4Interest-degree score value product be 8.
Wherein, the event pair of user A<E2、E8>In, the purchase events of webpage project y execute number of times and are
0, i.e., there are no the alternative events, the interest-degree score value of purchase events is 0, and therefore product is also 0,
The data can be rejected when calculating, save data volume and amount of calculation.
The interest-degree of the identical event centering alternative events and targeted interactive event that calculate multiple users divides
The sum of products of value.
Calculate square sum of the interest-degree score value of the alternative events of the identical event centering of multiple users.
Calculate the targeted interactive event of the identical event centering of multiple users interest-degree score value square
Sum.
Calculate the phase between the targeted interactive event of the identical event centering of multiple users and alternative events
Guan Du.
Such that it is able to obtain event pair<E5、E4>Middle alternative events E5With targeted interactive event E4Correlation
Degree about 0.88, event pair<E2、E8>Middle alternative events E2With targeted interactive event E8The degree of correlation about
For 0.58, event pair<E1、E4>Middle alternative events E1With targeted interactive event E4The degree of correlation be about 0.86.
In practical application, there is a kind of situation is that user only there occurs alternative events, for this feelings
Condition, when alternative events for capturing user's generation, is searched this alternative events and is handed over multiple targets
The mutual degree of correlation of event, extracts the corresponding net of targeted interactive event that the degree of correlation meets in the first preset range
Page project is used as target web project.
For example, when user D has carried out collection event E for webpage project y5, find E4With E5
There is degree of correlation S54, and S54=0.88, belong to default SijIn the range of the degree of correlation of ∈ [0.8,1],
Extract E4Corresponding webpage project x, and recommend active user.
When user E has carried out collection event E for webpage project x2, find E8With E2Exist related
Degree S28, but S28=0.58, it is not belonging to default SijIn the range of the degree of correlation of ∈ [0.8,1], can not be by
E8Corresponding webpage project y recommends active user.
When another kind of situation is that active user there occurs multiple alternative events, can be for each target interaction
Event, searches the targeted interactive event degree of correlation respectively with multiple alternative events;According to the target
The alternative events degree of correlation respectively with each alternative events, calculate the mark targeted interactive event with multiple
The recommendation degree of the correlation of alternative events;Extract the target interaction thing that recommendation degree meets in the second preset range
The corresponding webpage project of part is used as target web project.
For example, user D has carried out the browsing event E of 1 time for webpage project x1, to webpage project y
The browsing event E of 1 time is carried out5;Find E1With E4There is degree of correlation S14, and S14=0.86,
Find E5With E4There is degree of correlation S54, and S54=0.88, then recommend webpage project x (to hand over to user D
Mutual event E4Corresponding webpage project) recommendation degree be 1*0.86+1*0.88=1.74, belong to default
RujIn the range of the recommendation degree of ∈ [0.8 ,+∞], so using webpage project x as target web project recommendation
To active user.
In addition to setting the mode of recommendation degree threshold value, it is also possible to recommended by the sequence of recommendation degree, for example,
When the targeted interactive event that can recommend user D has 10, can be according to recommendation degree from high to low
Sequence, using corresponding for the sequence targeted interactive event of first 3 webpage project as target web project recommendation
To user D.
It should be noted that for embodiment of the method, in order to be briefly described, therefore which is all expressed as one it is
The combination of actions of row, but those skilled in the art should know, and the embodiment of the present application is not by described
Sequence of movement restriction because according to the embodiment of the present application, some steps can using other orders or
Person is carried out simultaneously.Secondly, those skilled in the art should also know, embodiment described in this description
Preferred embodiment is belonged to, necessary to involved action not necessarily the embodiment of the present application.
With reference to Fig. 3, show that a kind of webpage project recommendation device based on alternative events of the application is implemented
The structured flowchart of example, specifically can include as lower module:
Degree of correlation generation module 301, for being directed to the friendship of the alternative events of webpage project according to multiple users
Mutual information, generates the degree of correlation of the targeted interactive event in the alternative events and the alternative events.
Used as the preferred exemplary of the embodiment of the present invention, the degree of correlation generation module 401 can include following
Submodule:
Interest-degree score value generates submodule, for generating the friendship according to the event attribute of the alternative events
The interest-degree score value of mutual event.
Used as the preferred exemplary of the embodiment of the present invention, the alternative events include to browse thing to webpage project
Part, and/or collection event, and/or interpolation shopping cart event, and/or purchase events, the interest-degree divide
Value generates submodule can include following subelement:
Interest-degree score value M generates subelement, for the browsing event, the collection event, institute
State shopping cart event of adding, the alternative events in the purchase events and execute taking advantage of for times N and preset parameter
Product is M as the interest-degree score value of the then alternative events.
Relatedness computation submodule, for calculating the interaction according to the interest-degree score value of the alternative events
The degree of correlation of event and the targeted interactive event.
Used as the preferred exemplary one of the embodiment of the present invention, the relatedness computation submodule can include following
Subelement:
Event to constitute subelement, for the alternative events of same user are succeeded with targeted interactive event group
Part pair, and calculate taking advantage of for the alternative events of the event centering and the interest-degree score value of targeted interactive event
Product.
Relatedness computation subelement, for according to the alternative events of the identical event centering of multiple users
Interest-degree score value, the interest-degree score value of targeted interactive event and alternative events and targeted interactive event
The product of interest-degree score value, the alternative events for calculating the event centering are related to hand over targeted interactive event
Degree.
Used as the preferred exemplary of the embodiment of the present invention, the relatedness computation subelement can be specifically for:
For the identical event pair of multiple users, the interest of the alternative events of identical event centering is calculated
Square sum, square sum of the interest-degree score value of targeted interactive event of degree score value and alternative events
With the sum of products of the interest-degree score value of targeted interactive event, and the interaction of the event centering is correspondingly formed
Event multi-C vector value corresponding with targeted interactive event difference;Calculate remaining between the multi-C vector value
String value as the degree of correlation.
Used as the preferred exemplary two of the embodiment of the present invention, the relatedness computation submodule can include following
Subelement:
Interest-degree score value collection is combined into subelement, for the alternative events of same user are interacted thing with target
Part constitutes event pair, and interacts thing using the alternative events of the identical event centering of multiple users with target
The interest-degree score value of part separately constitutes the set of alternative events interest-degree score value and targeted interactive event interest-degree
Score value set.
The German several computation subunit of outstanding card, for calculate alternative events interest-degree score value set with described
The outstanding person of targeted interactive event interest-degree score value set blocks German number and as the degree of correlation.
Extraction module 302, for for active user alternative events, according to the alternative events with many
The degree of correlation of individual targeted interactive event extracts the corresponding target web project of at least one targeted interactive event.
As the preferred exemplary one of the embodiment of the present invention, when the active user there occurs alternative events
When, the extraction module 302 can include following submodule:
First degree of correlation searches submodule, for working as alternative events for capturing user generation,
Search the degree of correlation of one alternative events and multiple targeted interactive event.
First extracting sub-module, meets the targeted interactive event in the first preset range for extracting the degree of correlation
Corresponding webpage project is used as target web project.
As the preferred exemplary two of the embodiment of the present invention, when the active user there occurs multiple alternative events
When, the extraction module 302 can include following submodule:
Second degree of correlation searches submodule, for being directed to each targeted interactive event, searches the target and hands over
The mutual event degree of correlation respectively with multiple alternative events.
Recommendation degree calculating sub module, for according to the targeted interactive event respectively with each alternative events
The degree of correlation, calculates the recommendation degree of the mark targeted interactive event and the correlation of multiple alternative events.
Second extracting sub-module, meets the targeted interactive event in the second preset range for extracting recommendation degree
Corresponding webpage project is used as target web project.
Recommending module 303, for giving the active user by the target web project recommendation.
The webpage project can include that trading object, and/or video, and/or audio frequency, and/or electronics are read
Thing.
Webpage project is combined together to form by the device of the application with the interbehavior for webpage project
Alternative events, carry out relatedness computation based on alternative events, it is to avoid because user does not provide evaluation point
And lead to not carry out the problem of effective webpage project recommendation;And, using the friendship for webpage project
Mutually event can will meet the webpage item of user interest objective, exactly as the basis of relatedness computation
Mesh recommends user;Further, targeted interactive event is only calculated in relatedness computation with alternative events
The degree of correlation, reduces the data volume in relatedness computation, saves computing resource and the storage money of server
Source.
For device embodiment, due to itself and embodiment of the method basic simlarity, so the comparison of description
Simply, related part is illustrated referring to the part of embodiment of the method.
Each embodiment in this specification is all described by the way of going forward one by one, and each embodiment is stressed
Be all difference with other embodiment, between each embodiment identical similar part mutually referring to
?.
Those skilled in the art are it should be appreciated that the embodiment of the embodiment of the present application can be provided as method, dress
Put or computer program.Therefore, the embodiment of the present application can using complete hardware embodiment, completely
Software implementation or with reference to software and hardware in terms of embodiment form.And, the embodiment of the present application
Can adopt and can be situated between with storage in one or more computers for wherein including computer usable program code
The upper computer journey that implements of matter (including but not limited to magnetic disc store, CD-ROM, optical memory etc.)
The form of sequence product.
In a typical configuration, the computer equipment includes one or more processors
(CPU), input/output interface, network interface and internal memory.Internal memory potentially includes computer-readable medium
In volatile memory, the shape such as random access memory (RAM) and/or Nonvolatile memory
Formula, such as read-only storage (ROM) or flash memory (flash RAM).Internal memory is computer-readable medium
Example.Computer-readable medium includes permanent and non-permanent, removable and non-removable media
Information Store can be realized by any method or technique.Information can be computer-readable instruction,
Data structure, the module of program or other data.The example of the storage medium of computer includes, but
It is not limited to phase transition internal memory (PRAM), static RAM (SRAM), dynamic random to deposit
Access to memory (DRAM), other kinds of random access memory (RAM), read-only storage
(ROM), Electrically Erasable Read Only Memory (EEPROM), fast flash memory bank or other in
Deposit technology, read-only optical disc read-only storage (CD-ROM), digital versatile disc (DVD) or other
Optical storage, magnetic cassette tape, tape magnetic rigid disk storage other magnetic storage apparatus or any its
His non-transmission medium, can be used to store the information that can be accessed by a computing device.According to herein
Define, computer-readable medium does not include the computer readable media (transitory media) of non-standing,
Data-signal and carrier wave as modulation.
The embodiment of the present application be with reference to according to the method for the embodiment of the present application, terminal device (system) and meter
The flow chart of calculation machine program product and/or block diagram are describing.It should be understood that can be by computer program instructions
Each flow process and/or square frame and flow chart and/or square frame in flowchart and/or block diagram
The flow process of in figure and/or the combination of square frame.Can provide these computer program instructions to all-purpose computer,
The processor of special-purpose computer, Embedded Processor or other programmable data processing terminal equipments is to produce
One machine so that by the computing device of computer or other programmable data processing terminal equipments
Instruction produce for realizing in one flow process of flow chart or one square frame of multiple flow processs and/or block diagram or
The device of the function of specifying in multiple square frames.
These computer program instructions may be alternatively stored in and computer or other programmable datas can be guided to process
In the computer-readable memory that terminal device is worked in a specific way so that be stored in the computer-readable
Instruction in memory is produced and includes the manufacture of command device, and the command device is realized in flow chart one
The function of specifying in flow process or one square frame of multiple flow processs and/or block diagram or multiple square frames.
These computer program instructions can also be loaded into computer or other programmable data processing terminals set
Standby upper so as to execute series of operation steps on computer or other programmable terminal equipments in terms of producing
The process that calculation machine is realized, the instruction so as to execute on computer or other programmable terminal equipments provide use
In realization in one flow process of flow chart or one square frame of multiple flow processs and/or block diagram or multiple square frames
The step of function of specifying.
The preferred embodiment of the embodiment of the present application is although had been described for, but those skilled in the art are once
Basic creative concept is known, then other change and modification can be made to these embodiments.So,
Claims are intended to be construed to include preferred embodiment and fall into the institute of the embodiment of the present application scope
Have altered and change.
Finally, in addition it is also necessary to explanation, herein, such as first and second or the like relational terms
Be used merely to an entity or operation is made a distinction with another entity or operation, and not necessarily require
Or imply between these entities or operation, there is any this actual relation or order.And, art
Language " including ", "comprising" or its any other variant are intended to including for nonexcludability, so that
A series of process, method, article or terminal device including key elements not only includes those key elements, and
Also include other key elements being not expressly set out, or also include for this process, method, article or
The intrinsic key element of person's terminal device.In the absence of more restrictions, by sentence " including one
Individual ... " key element that limits, it is not excluded that at the process, method, article or the end that include the key element
Also there is other identical element in end equipment.
Above to a kind of webpage item recommendation method based on alternative events provided herein and one kind
Based on the webpage project recommendation device of alternative events, it is described in detail, used herein specifically individual
Example is set forth to the principle of the application and embodiment, and the explanation of above example is only intended to help
Understand the present processes and its core concept;Simultaneously for one of ordinary skill in the art, foundation
The thought of the application, all will change in specific embodiments and applications, in sum,
This specification content should not be construed as the restriction to the application.
Claims (18)
1. a kind of webpage item recommendation method based on alternative events, it is characterised in that include:
According to multiple users for the interactive information of the alternative events of webpage project, the alternative events are generated
The degree of correlation with the targeted interactive event in the alternative events;
For the alternative events of active user, according to the phase of the alternative events and multiple targeted interactive event
Guan Du extracts the corresponding target web project of at least one targeted interactive event;
The active user is given in the target web project recommendation.
2. method according to claim 1, it is characterised in that described be directed to according to multiple users
The interactive information of the alternative events of webpage project, generates the mesh in the alternative events and the alternative events
The step of degree of correlation of mark alternative events, includes:
The interest-degree score value of the alternative events is generated according to the event attribute of the alternative events;
The alternative events are calculated according to the interest-degree score value of the alternative events thing is interacted with the target
The degree of correlation of part.
3. method according to claim 2, it is characterised in that the alternative events are included to net
The browsing event of page project, and/or collection event, and/or interpolation shopping cart event, and/or purchase events,
Described generate the alternative events according to the event attribute of alternative events interest-degree score value the step of include:
With the browsing event, the collection event, the interpolation shopping cart event, the purchase events
In alternative events execute the interest-degree of the product as the then alternative events of times N and preset parameter
Score value is M.
4. method according to claim 2, it is characterised in that described according to the emerging of alternative events
The step of interesting degree score value calculates the degree of correlation of the alternative events and the targeted interactive event includes:
The alternative events of same user are constituted event pair with targeted interactive event, and calculates the event pair
In alternative events and targeted interactive event interest-degree score value product;
Interest-degree score value according to the alternative events of the identical event centering of multiple users, target interaction thing
The interest-degree score value of part and alternative events and the product of the interest-degree score value of targeted interactive event, calculate
The alternative events of the event centering and the degree of correlation for handing over targeted interactive event.
5. method according to claim 4, it is characterised in that the phase according to multiple users
With the interest-degree score value of alternative events of event centering, the interest-degree score value of targeted interactive event and
Alternative events and the product of the interest-degree score value of targeted interactive event, calculate the interactive thing of the event centering
Part is specially with the degree of correlation for handing over targeted interactive event:
For the identical event pair of multiple users, the interest of the alternative events of identical event centering is calculated
Square sum, square sum of the interest-degree score value of targeted interactive event of degree score value and alternative events
With the sum of products of the interest-degree score value of targeted interactive event, and the interaction of the event centering is correspondingly formed
Event multi-C vector value corresponding with targeted interactive event difference;
Calculate cosine value between the multi-C vector value and as the degree of correlation.
6. method according to claim 2, it is characterised in that described according to the alternative events
Interest-degree score value calculate the alternative events and the targeted interactive event the degree of correlation the step of include:
The alternative events of same user are constituted event pair with targeted interactive event, and using multiple users'
The alternative events of identical event centering are separately constituted with the interest-degree score value of targeted interactive event and interact thing
The set of part interest-degree score value and targeted interactive event interest-degree score value set;
Alternative events interest-degree score value set is calculated with the targeted interactive event interest-degree score value collection
The outstanding person of conjunction blocks German number and as the degree of correlation.
7. method according to claim 1, it is characterised in that when the active user there occurs
During one alternative events, the alternative events for active user, according to the alternative events with multiple
The degree of correlation of targeted interactive event extracts the corresponding target web project of at least one targeted interactive event
Step includes:
When alternative events for capturing user generation, search one alternative events with multiple
The degree of correlation of targeted interactive event;
Extract the degree of correlation and meet the corresponding webpage project conduct of targeted interactive event in the first preset range
Target web project.
8. method according to claim 1, it is characterised in that when the active user there occurs
During multiple alternative events, the alternative events for active user, according to the alternative events with multiple
The degree of correlation of targeted interactive event extracts the corresponding target web project of at least one targeted interactive event
Step includes:
For each targeted interactive event, search the targeted interactive event respectively with multiple alternative events
The degree of correlation;
According to the targeted interactive event degree of correlation respectively with each alternative events, the mark mesh is calculated
Mark the recommendation degree of alternative events and the correlation of multiple alternative events;
Extract recommendation degree and meet the corresponding webpage project conduct of targeted interactive event in the second preset range
Target web project.
9. method according to claim 1, it is characterised in that the webpage project includes transaction
Object, and/or video, and/or audio frequency, and/or electron reading.
10. a kind of webpage project recommendation device based on alternative events, it is characterised in that include:
Degree of correlation generation module, for believing for the interaction of the alternative events of webpage project according to multiple users
Breath, generates the degree of correlation of the targeted interactive event in the alternative events and the alternative events;
Extraction module, for the alternative events for active user, according to the alternative events and multiple mesh
The degree of correlation of mark alternative events extracts the corresponding target web project of at least one targeted interactive event;
Recommending module, for giving the active user by the target web project recommendation.
11. devices according to claim 10, it is characterised in that the degree of correlation generation module
Including:
Interest-degree score value generates submodule, for generating the friendship according to the event attribute of the alternative events
The interest-degree score value of mutual event;
Relatedness computation submodule, for calculating the interaction according to the interest-degree score value of the alternative events
The degree of correlation of event and the targeted interactive event.
12. devices according to claim 11, it is characterised in that it is right that the alternative events include
The browsing event of webpage project, and/or collection event, and/or interpolation shopping cart event, and/or purchase thing
Part, the interest-degree score value generates submodule to be included:
Interest-degree score value M generates subelement, for the browsing event, the collection event, institute
State shopping cart event of adding, the alternative events in the purchase events and execute taking advantage of for times N and preset parameter
Product is M as the interest-degree score value of the then alternative events.
13. devices according to claim 11, it is characterised in that the relatedness computation submodule
Block includes:
Event to constitute subelement, for the alternative events of same user are succeeded with targeted interactive event group
Part pair, and calculate taking advantage of for the alternative events of the event centering and the interest-degree score value of targeted interactive event
Product;
Relatedness computation subelement, for according to the alternative events of the identical event centering of multiple users
Interest-degree score value, the interest-degree score value of targeted interactive event and alternative events and targeted interactive event
The product of interest-degree score value, the alternative events for calculating the event centering are related to hand over targeted interactive event
Degree.
14. devices according to claim 13, it is characterised in that relatedness computation is single
Unit specifically for:
For the identical event pair of multiple users, the interest of the alternative events of identical event centering is calculated
Square sum, square sum of the interest-degree score value of targeted interactive event of degree score value and alternative events
With the sum of products of the interest-degree score value of targeted interactive event, and the interaction of the event centering is correspondingly formed
Event multi-C vector value corresponding with targeted interactive event difference;
Calculate cosine value between the multi-C vector value and as the degree of correlation.
15. devices according to claim 11, it is characterised in that the relatedness computation submodule
Block includes:
Interest-degree score value collection is combined into subelement, for the alternative events of same user are interacted thing with target
Part constitutes event pair, and interacts thing using the alternative events of the identical event centering of multiple users with target
The interest-degree score value of part separately constitutes the set of alternative events interest-degree score value and targeted interactive event interest-degree
Score value set;
The German several computation subunit of outstanding card, for calculate alternative events interest-degree score value set with described
The outstanding person of targeted interactive event interest-degree score value set blocks German number and as the degree of correlation.
16. devices according to claim 10, it is characterised in that when the active user occurs
During one alternative events, the extraction module includes:
First degree of correlation searches submodule, for working as alternative events for capturing user generation,
Search the degree of correlation of one alternative events and multiple targeted interactive event;
First extracting sub-module, meets the targeted interactive event in the first preset range for extracting the degree of correlation
Corresponding webpage project is used as target web project.
17. devices according to claim 10, it is characterised in that when the active user occurs
During multiple alternative events, the extraction module includes:
Second degree of correlation searches submodule, for being directed to each targeted interactive event, searches the target and hands over
The mutual event degree of correlation respectively with multiple alternative events;
Recommendation degree calculating sub module, for according to the targeted interactive event respectively with each alternative events
The degree of correlation, calculates the recommendation degree of the mark targeted interactive event and the correlation of multiple alternative events;
Second extracting sub-module, meets the targeted interactive event in the second preset range for extracting recommendation degree
Corresponding webpage project is used as target web project.
18. devices according to claim 10, it is characterised in that the webpage project includes to hand over
Easy object, and/or video, and/or audio frequency, and/or electron reading.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510557884.5A CN106484747A (en) | 2015-09-02 | 2015-09-02 | A kind of webpage item recommendation method based on alternative events and device |
PCT/CN2016/096583 WO2017036333A1 (en) | 2015-09-02 | 2016-08-24 | Interaction event-based webpage item recommendation method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510557884.5A CN106484747A (en) | 2015-09-02 | 2015-09-02 | A kind of webpage item recommendation method based on alternative events and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106484747A true CN106484747A (en) | 2017-03-08 |
Family
ID=58186673
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510557884.5A Pending CN106484747A (en) | 2015-09-02 | 2015-09-02 | A kind of webpage item recommendation method based on alternative events and device |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106484747A (en) |
WO (1) | WO2017036333A1 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110647676B (en) * | 2019-08-14 | 2023-04-11 | 平安科技(深圳)有限公司 | Interest attribute mining method and device based on big data and computer equipment |
CN113177184A (en) * | 2021-04-22 | 2021-07-27 | 武汉理工大学 | Building material equipment manufacturing enterprise supplier selection method, equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101124575A (en) * | 2004-02-26 | 2008-02-13 | 雅虎公司 | Method and system for generating recommendations |
CN103810030A (en) * | 2014-02-20 | 2014-05-21 | 北京奇虎科技有限公司 | Application recommendation method, device and system based on mobile terminal application market |
CN104281718A (en) * | 2014-11-04 | 2015-01-14 | 深圳市英威诺科技有限公司 | Method for performing data mining intelligent recommendation based on user group behavior |
CN104765751A (en) * | 2014-01-07 | 2015-07-08 | 腾讯科技(深圳)有限公司 | Application recommendation method and device |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102456056B (en) * | 2010-11-01 | 2013-07-03 | 阿里巴巴集团控股有限公司 | Information output method and information output device |
CN102866992B (en) * | 2011-07-04 | 2015-12-02 | 阿里巴巴集团控股有限公司 | A kind of method and device showing product information in webpage |
-
2015
- 2015-09-02 CN CN201510557884.5A patent/CN106484747A/en active Pending
-
2016
- 2016-08-24 WO PCT/CN2016/096583 patent/WO2017036333A1/en active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101124575A (en) * | 2004-02-26 | 2008-02-13 | 雅虎公司 | Method and system for generating recommendations |
CN104765751A (en) * | 2014-01-07 | 2015-07-08 | 腾讯科技(深圳)有限公司 | Application recommendation method and device |
CN103810030A (en) * | 2014-02-20 | 2014-05-21 | 北京奇虎科技有限公司 | Application recommendation method, device and system based on mobile terminal application market |
CN104281718A (en) * | 2014-11-04 | 2015-01-14 | 深圳市英威诺科技有限公司 | Method for performing data mining intelligent recommendation based on user group behavior |
Also Published As
Publication number | Publication date |
---|---|
WO2017036333A1 (en) | 2017-03-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9870408B2 (en) | Search method and system | |
JP6422617B2 (en) | Network access operation identification program, server, and storage medium | |
US8838583B1 (en) | Diversity within search results | |
TWI512653B (en) | Information providing method and apparatus, method and apparatus for determining the degree of comprehensive relevance | |
CN104699725B (en) | data search processing method and system | |
JP5897019B2 (en) | Method and apparatus for determining linked list of candidate products | |
CN106708844A (en) | User group partitioning method and device | |
US9460463B2 (en) | Method, web server and web browser of providing information | |
CN109816482B (en) | Knowledge graph construction method, device and equipment of e-commerce platform and storage medium | |
CN107066476A (en) | A kind of real-time recommendation method based on article similarity | |
CN109447713A (en) | A kind of recommended method and device of knowledge based map | |
EP2641161A1 (en) | Transmitting product information | |
CN103345698A (en) | Personalized recommendation method based on cloud processing mode and applied in e-business environment | |
CN110335123B (en) | Commodity recommendation method, system, computer readable medium and device based on social e-commerce platform | |
CN106251168A (en) | Information-pushing method and system | |
CN102298650B (en) | Distributed recommendation method of massive digital information | |
CN105023178B (en) | A kind of electronic commerce recommending method based on ontology | |
TW201301064A (en) | Search Method and Apparatus | |
CN103092861A (en) | Method and system for selecting commodity representative picture | |
CN104053023A (en) | Method and device for determining video similarity | |
CN103530416A (en) | Project data forecasting grading library generating and project data pushing method and project data forecasting grading library generating and project data pushing system | |
CN103365842B (en) | A kind of page browsing recommends method and device | |
CN106919582A (en) | The association of network articles and related information statistical method and device | |
CN103902549A (en) | Search data sorting method and device and data searching method and device | |
CN110309410A (en) | A kind of information recommended method, platform and computer readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170308 |