CN108198019A - Item recommendation method and device, storage medium, electronic equipment - Google Patents
Item recommendation method and device, storage medium, electronic equipment Download PDFInfo
- Publication number
- CN108198019A CN108198019A CN201711450258.1A CN201711450258A CN108198019A CN 108198019 A CN108198019 A CN 108198019A CN 201711450258 A CN201711450258 A CN 201711450258A CN 108198019 A CN108198019 A CN 108198019A
- Authority
- CN
- China
- Prior art keywords
- user
- preference
- time
- real
- article
- 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
Abstract
Embodiments of the present invention provide a kind of item recommendation method and device based on user preference, belong to technical field of data processing, this method includes:The first default processing is carried out to user's history behavioral data and obtains long-term preference in user;To user, real-time behavioral data carries out the second default processing and obtains the real-time preference of user;Corresponding article is recommended for user according to preference long-term in the user and the real-time preference of user.This method preferably can recommend corresponding article according to the preference of user and demand for user, increase the fineness of article recommendation results, user is allowd to select the article for meeting oneself demand within the shortest time, the spending limit of user can also be increased simultaneously by reducing user and selecting time of article.
Description
Technical field
Embodiments of the present invention are related to data processing field, more specifically, embodiments of the present invention are related to based on use
The item recommendation method of family preference, article recommendation apparatus, storage medium and electronic equipment based on user preference.
Background technology
Background that this section is intended to provide an explanation of the embodiments of the present invention set forth in the claims or context.Herein
Description recognizes it is the prior art not because not being included in this part.
With the rapid development of e-commerce, more and more consumer's selections are purchased by shopping at network to substitute entity
Object.In order to cause consumer finds most of the article met, major shopping website within the shortest time all to pass through
Recommended according to the previous purchasing habits of user for user.
In existing collaborative filtering item recommendation method, mainly recommend some and the object liked before them to user
Article as condition;But the algorithm does not utilize the contents attribute of article to calculate the similarity between article, it mainly passes through
Analyze the similarity between the behavior record calculating article of user.The algorithm thinks, article A and article B have very big similar
Degree is because the user of article A is liked mostly also to like article B, and the collaborative filtering based on article is broadly divided into two steps:First,
Calculate the similarity between article;2nd, recommendation list is generated to user according to the historical behavior of the similarity of article and user.
Invention content
But in some technologies, on the one hand, user preference is not analyzed so that the recommendation results of article are not
It is enough fine, it is impossible to the correctly hobby of reaction user so that user can not select within the shortest time to be met oneself and need
The article asked;On the other hand, recommend scene excessively single so that user cannot can also be seen that while a kind of article is selected
The article that other needs are bought reduces the consumption number of user;Another further aspect, it is impossible to which personalization is done to the article of given set
Recommend, reduce the interest for recommending the page while the wish for reducing user's purchase.
It therefore in the prior art, cannot be when shorter by user's recommendation article of the method that collaborative filtering article is recommended
Interior to reach preferable consumption purpose, this is very bothersome process.
Thus, it is also very desirable to a kind of improved item recommendation method based on user preference, the article based on user preference
Recommendation apparatus, storage medium and electronic equipment, so that user can timely find according to recommendation results meets oneself demand
Article, and increase spending limit.
In the present context, embodiments of the present invention be intended to provide a kind of item recommendation method based on user preference,
Article recommendation apparatus, storage medium and electronic equipment based on user preference.
According to one aspect of the disclosure, a kind of item recommendation method based on user preference is provided, including:
The first default processing is carried out to user's history behavioral data and obtains long-term preference in user;
To user, real-time behavioral data carries out the second default processing and obtains the real-time preference of user;
Corresponding article is recommended for user according to preference long-term in the user and the real-time preference of user.
In a kind of exemplary embodiment of the disclosure, it is according to preference long-term in the user and the real-time preference of user
User recommends corresponding article to include:
Processing is preset to the user real-time preference progress third and obtains user view;
User is obtained according to preference long-term in the user, the real-time preference of user and user view and integrates preference;
Preference is integrated according to the user and recommends article corresponding with user synthesis preference for user.
In a kind of exemplary embodiment of the disclosure, the behavioral data includes navigation patterns data, collection behavior number
According to, plus purchase behavioral data, buying behavior data and exposure behavioral data in it is a variety of.
It is default to user's history behavioral data progress first to handle in a kind of exemplary embodiment of the disclosure
Into user, long-term preference includes:
Setup time attenuation function, and during according to the generations of the time attenuation function and each historical behavior data
Between and frequency calculate the first user preference value of each historical behavior data;
For each first weighted value of historical behavior data configuration, and according to each historical behavior data corresponding first
User preference value and the first weighted value calculate long-term preference in the user.
It is default to the user real-time behavioral data progress second to handle in a kind of exemplary embodiment of the disclosure
Include to the real-time preference of user:
It is calculated according to the time of origin and frequency of the time attenuation function and each real-time behavioral data
The second user preference value of each real-time behavioral data;
The second weighted value is configured, and according to each real-time behavioral data corresponding second for each real-time behavioral data
User preference value and the second weighted value calculate the real-time preference of user.
In a kind of exemplary embodiment of the disclosure, the user preference includes user's classification preference and user brand
Preference;
Third is carried out to the real-time preference of the user and presets processing and obtain user view to include:
Configuration information entropy function, and the use is predicted according to described information entropy function and the real-time classification preference of the user
Family is intended to.
In a kind of exemplary embodiment of the disclosure, according to preference long-term in the user, the real-time preference of user with
And after user view obtains user's synthesis preference, the item recommendation method based on user preference further includes:
Preference long-term in the user, the real-time preference of user and user view are stored into distributed caching cluster.
In a kind of exemplary embodiment of the disclosure, carry out the second default processing in behavioral data real-time to user and obtain
After the real-time preference of user, the item recommendation method based on user preference further includes:
Preference long-term in the user and the real-time preference of user are normalized.
In a kind of exemplary embodiment of the disclosure, it is according to preference long-term in the user and the real-time preference of user
User recommends corresponding article to further include:
According to different type of recommendation, for preference long-term in the user after normalized and the real-time preference configuration of user
Proportionality coefficient;
It is calculated and used according to the proportionality coefficient of preference long-term in the user after the normalized and the real-time preference of user
Family entirety preference;
Corresponding article is recommended for the user according to user's entirety preference.
In a kind of exemplary embodiment of the disclosure, the attenuation function is:
Wherein, β is time constant;tuiGeneration time for user behavior data;T is to calculate preference
Time.
According to one aspect of the disclosure, a kind of article recommendation apparatus based on user preference is provided, including:
First processing module, it is long-term partially in user for being obtained to the first default processing of user's history behavioral data progress
It is good;
Second processing module carries out the second default processing for behavioral data real-time to user and obtains the real-time preference of user;
Article recommending module, for recommending to correspond to for user according to preference long-term in the user and the real-time preference of user
Article.
According to one aspect of the disclosure, a kind of storage medium is provided, is stored thereon with computer program, the computer
The item recommendation method based on user preference described in above-mentioned any one is realized when program is executed by processor.
According to one aspect of the disclosure, a kind of electronic equipment is provided, including:
Processor;And
Memory, for storing the executable instruction of the processor;
Wherein, the processor is configured to perform the base described in above-mentioned any one via the executable instruction is performed
In the item recommendation method of user preference.
It is pushed away according to the item recommendation method based on user preference of embodiment of the present invention and the article based on user preference
Device is recommended, by the medium-term and long-term preference for calculating user and real-time preference, then according to the medium-term and long-term preference of user and in real time
Preference recommends corresponding article for user, preferably can recommend corresponding object for user according to the preference of user and demand
Product increase the fineness of article recommendation results so that user can select within the shortest time meets oneself demand
Article, the spending limit of user can also be increased simultaneously by reducing user and selecting time of article.
Description of the drawings
Detailed description below, above-mentioned and other mesh of exemplary embodiment of the invention are read by reference to attached drawing
, feature and advantage will become prone to understand.In the accompanying drawings, if showing the present invention's by way of example rather than limitation
Dry embodiment, wherein:
Fig. 1 schematically shows the flow of the item recommendation method based on user preference according to embodiment of the present invention
Figure;
Fig. 2 schematically shows the frames of the item recommendation method based on user preference according to embodiment of the present invention
Figure;
Fig. 3, which is schematically shown, carries out user's history behavioral data according to embodiment of the present invention at the first default place
Reason obtains the method flow diagram of long-term preference in user;
Fig. 4 schematically shows according to embodiment of the present invention, to user, real-time behavioral data carries out the second default place
Reason obtains the method flow diagram of the real-time preference of user;
Fig. 5 schematically shows real according to preference long-term in the user and user according to embodiment of the present invention
When preference recommend the method flow diagram of corresponding article for user;
Fig. 6 schematically show according to embodiment of the present invention it is another according to preference long-term in the user and
The real-time preference of user recommends the method flow diagram of corresponding article for user;
Fig. 7 schematically shows the frame of the article recommendation apparatus based on user preference according to embodiment of the present invention
Figure;
Fig. 8 schematically shows the schematic diagram of the storage medium according to embodiment of the present invention;
Fig. 9 schematically shows the block diagram figure according to the electronic equipment of embodiment of the present invention.
In the accompanying drawings, identical or corresponding label represents identical or corresponding part.
Specific embodiment
The principle and spirit of the invention are described below with reference to several illustrative embodiments.It should be appreciated that provide this
A little embodiments are not with any just for the sake of better understood when those skilled in the art and then realize the present invention
Mode limits the scope of the invention.On the contrary, these embodiments are provided so that the present invention is more thorough and complete, and energy
It enough will fully convey the scope of the invention to those skilled in the art.
One skilled in the art will appreciate that embodiments of the present invention can be implemented as a kind of system, device, equipment, method
Or computer program product.Therefore, the present invention can be with specific implementation is as follows, i.e.,:Complete hardware, complete software
The form that (including firmware, resident software, microcode etc.) or hardware and software combine.
According to the embodiment of the present invention, it is proposed that a kind of item recommendation method based on user preference, inclined based on user
Good article recommendation apparatus, storage medium and electronic equipment.
Herein, any number of elements in attached drawing is used to example and unrestricted and any name is only used for
It distinguishes, without any restrictions meaning.
Below with reference to several representative embodiments of the present invention, the principle and spirit of the invention are illustrated in detail.
Summary of the invention
The inventors discovered that in existing article suggested design, the method for using collaborative filtering is recommended and is used for user
The mode of the similar article of the article liked before family is more single, can not be that more fine recommendation results and not are presented in user
The hobby of user can correctly be reacted;In addition, since user had bought article A, it will not be again in the short time
Purchase has the article B of very big similarity with article A, this may cause user that can generate what is more detested to the article B of recommendation
At heart, user's shopping experience is reduced, and then user cannot be made to generate the wish bought well.Further, since to given collection
The article of conjunction does personalized recommendation, therefore reduces the interest for recommending the page while the wish for reducing user's purchase.
In view of the above, basic thought of the invention is:On the one hand, by calculate user medium-term and long-term preference and
Then real-time preference recommends corresponding article according to the medium-term and long-term preference of user and real-time preference for user, can be better
Corresponding article is recommended for user according to the preference of user and demand, increases the fineness of article recommendation results so that is used
Family can select the article for meeting oneself demand within the shortest time, reduce user and select the time of article while also may be used
To increase the spending limit of user;On the other hand, user view is predicted, can is more accurately user recommended user
Required article improves the shopping experience of user;In another aspect, corresponding object is recommended for user according to different type of recommendation
Product can do personalized recommendation to the article or brand and classification etc. of given set, increase the entertaining of user's shopping
Property.
After the basic principle for describing the present invention, lower mask body introduces the various nonrestrictive embodiment party of the present invention
Formula.
Illustrative methods
The article recommendation side based on user preference according to exemplary embodiment of the invention is described with reference to Fig. 1
Method.
It may comprise steps of refering to what is shown in Fig. 1, being somebody's turn to do the item recommendation method based on user preference:
Step S110. carries out user's history behavioral data the first default processing and obtains long-term preference in user.
To user, real-time the second default processing of behavioral data progress obtains the real-time preference of user to step S120..
Step S130. recommends corresponding article according to preference long-term in the user and the real-time preference of user for user.
In the item recommendation method based on user preference of exemplary embodiment of the invention, by calculating in user
Then long-term preference and real-time preference recommend corresponding object according to the medium-term and long-term preference of user and real-time preference for user
Product preferably can recommend corresponding article for user according to the preference of user and demand, increase article recommendation results
Fineness so that user can select the article for meeting oneself demand within the shortest time, reduce user and select article
Time can also increase the spending limit of user simultaneously.
In the following, by with reference to attached drawing in the item recommendation method based on user preference of exemplary embodiment of the invention
Each step carries out detailed explanation and explanation.
Refering to what is shown in Fig. 1, in step s 110, the first default processing is carried out to user's history behavioral data and is obtained in user
Long-term preference.
In this illustrative embodiments, user's history behavioral data can include user to the browsing (Click) of article,
Buy (Buy), collection (Follow) and the data that the behaviors such as shopping cart (Cart) is added to generate;Article can also be treated including user
The data that generate of other behaviors, such as can be exposure (Show) etc., this illustrative embodiment does not do this special limit
System;Wherein, user's history behavioral data can include user's behavioral data of nearly n days, n can including 3 days, 7 days, 15 days, 30
It or 60 days etc., or other data, such as can be 90 days etc., there is no special restriction on this for this example.Into
One step, refering to what is shown in Fig. 2, first, the historical behavior data of user are obtained from database 201;Then, pass through off-line calculation
Cluster 203 carries out the historical behavior data got the first default processing and obtains long-term preference in user;Wherein, off-line calculation
Cluster can include Hive clusters and Hadoop clusters, can also include other clusters, such as can be Kafka clusters etc.
Deng there is no special restriction on this for this example.
With continued reference to shown in Fig. 1, in the step s 120, to user, real-time the second default processing of behavioral data progress is used
The real-time preference in family.
In this illustrative embodiments, the real-time behavioral data of user can include user to the browsing (Click) of article,
Buy (Buy), collection (Follow) and the data that the behaviors such as shopping cart (Cart) is added to generate;Article can also be treated including user
The data that generate of other behaviors, such as can be exposure (Show) etc., this illustrative embodiment does not do this special limit
System.Further, with continued reference to shown in Fig. 2, first, the real-time behavioral data of user is obtained from database 201;Then, lead to
It crosses real-time computing cluster 205 and the real-time preference of user is obtained to the second default processing of real-time behavioral data progress got;Wherein,
Real-time computing cluster can include Kafka clusters or Hadoop clusters, can also include other clusters, such as can be Hive
Cluster etc., there is no special restriction on this for this example.
With continued reference to shown in Fig. 1, in step s 130, it is according to preference long-term in the user and the real-time preference of user
User recommends corresponding article.
In this illustrative embodiments, after long-term preference and real-time preference is obtained in above-mentioned user, using online
Computing cluster 207 handles long-term preference and real-time preference in user, and recommends to correspond to for user according to handling result
Article;Wherein, online computing cluster can include Kafka clusters or Hadoop clusters, can also include other clusters, example
Such as can be Hive clusters, there is no special restriction on this for this example.
Fig. 3 schematically shows a kind of handle default to user's history behavioral data progress first and obtains in user for a long time
The method flow diagram of preference.It is obtained in user for a long time refering to what is shown in Fig. 3, carrying out the first default processing to user's history behavioral data
Preference can include step S310 and step S320.
Refering to what is shown in Fig. 3, in step S310, setup time attenuation function, and according to the time attenuation function and
The time of origin and frequency of each historical behavior data calculate the first user preference of each historical behavior data
Value.
In this illustrative embodiments, first, setup time attenuation function, wherein, time attenuation function for example can be with
For:Wherein, β is time constant;tuiGeneration time for user behavior data (such as can be 2017-12-
20 20:08);T is that calculate time of preference (such as can be 2017-12-22 17:08);Secondly, when time attenuation function is matched
After the completion of putting, can respectively it be gone through according to time attenuation function and the calculating of the time of origin and frequency of each historical behavior data
First user preference value of history behavioral data.For example:
This is sentenced for user is to browsing (Click) behavior of article in user's history behavioral data, further to root
According to time attenuation function and navigation patterns time of origin and frequency calculate the first user preference value of navigation patterns into
Row explanation.For example, a certain user is 100 times to the navigation patterns number to article, then above-mentioned time attenuation function can be utilized
And the time of origin in 100 navigation patterns each time calculates the user preference value of navigation patterns each time;When each time
After the user preference value of navigation patterns calculates completion, the summation for calculating the user preference value of navigation patterns each time is browsed
The user preference value of behavior.
In addition, user to the purchase (Buy) of article, collection (Follow) and adds shopping cart in user's history behavioral data
(Cart) etc. behaviors user preference value is similar to the user preference value calculating method of browsing (Click) behavior of article with user,
It no longer repeats one by one herein.
It is each first weighted value of historical behavior data configuration in step s 320, and root with continued reference to shown in Fig. 3
Long-term preference in the user is calculated according to corresponding first user preference value of each historical behavior data and the first weighted value.
In this illustrative embodiments, as the corresponding user of each behavior in above-mentioned user's history behavioral data that respectively obtains
After preference value, it is also necessary to be each the first weighted value of historical behavior data configuration;Wherein, the configuration of weighted value can be according to each history
Influence of the behavioral data to preference long-term in user determines;If for example, number of visits and receipts of the user to a certain article
It is more to hide number, then illustrates that the wish that the user more likes or buys to the article is more, therefore can be navigation patterns
And heavier weighted value is configured in collection behavior;If a certain user had bought a certain article whithin a period of time, can
To illustrate that a period of time no longer needs to buy the article again the user recently, therefore can be that buying behavior configuration is lighter
Weighted value.
Further, in order to be divided the preference of user so that recommendation results are more accurate in more detail
Really, user preference can be divided into user's classification preference and user brand preference;Further, based on above-mentioned weighted value with
And user's classification preference for being worth to of the corresponding preference of each user behavior and user brand preference can be as follows:
In above formula,For classification preference value long-term in the user of the i-th class article;Use for the i-th class article
Long-term Brang Preference value in family;α1、α2、α3And α4For the corresponding weighted value of each historical behavior data.
Fig. 4 schematically shows a kind of the second default processing of behavioral data progress real-time to user and obtains the real-time preference of user
Method flow diagram.Refering to what is shown in Fig. 4, to user real-time behavioral data carry out the second default processing obtain the real-time preference of user can be with
Including step S410 and step S420.
Refering to what is shown in Fig. 4, in step S410, according to the time attenuation function and each real-time behavioral data
Time of origin and frequency calculate the second user preference value of each real-time behavioral data.
It in this illustrative embodiments, can be according to the generation of above-mentioned time attenuation function and each real-time behavioral data
Time and frequency calculate the second user preference value of each real-time behavioral data.For example:
This is sentenced for user is to collection (Follow) behavior of article in the real-time behavioral data of user, further to root
According to time attenuation function and collection behavior time of origin and frequency calculate navigation patterns the first user preference value into
Row explanation.For example, a certain user to being 50 times to the navigation patterns number of article, then can utilize above-mentioned time attenuation function with
And the time of origin in 50 collection behaviors each time calculates the user preference value for collecting behavior each time;It is collected when each time
After the user preference value of behavior calculates completion, the summation for calculating the user preference value of collection behavior each time obtains collection behavior
User preference value.
In addition, user is to the browsing (Click) of article, purchase (Buy), collection in user's history behavioral data
(Follow), add the behaviors user preference value such as shopping cart (Cart) and exposure (Show) and collection of the user to article
(Follow) the user preference value calculating method of behavior is similar, no longer repeats one by one herein.
With continued reference to shown in Fig. 4, in step S410, the second weighted value, and root is configured for each real-time behavioral data
The real-time preference of user is calculated according to the corresponding second user preference value of each real-time behavioral data and the second weighted value.
In this illustrative embodiments, as the corresponding user of each behavior in the real-time behavioral data of above-mentioned user that respectively obtains
After preference value, it is also necessary to the first weighted value be configured for each real-time behavioral data;Wherein, the configuration of weighted value can be according to each real-time
The influence of behavioral data real-time preference to user determines;If for example, number of visits and collection of the user to a certain article
Number is more, then illustrates that the wish that the user more likes or buys to the article is more, thus can be navigation patterns with
And heavier weighted value is configured in collection behavior;If a certain user was buying a certain article recently, it can be said that bright should
A period of time no longer needs to buy the article again user recently, therefore can be that lighter weighted value is configured in buying behavior.
Further, since the real-time preference of user can be divided into the real-time classification preference of user and the real-time brand of user is inclined
It is good;Therefore, based on above-mentioned weighted value and the real-time classification preference of the available user of the corresponding preference value of each user behavior with
And the real-time Brang Preference of user is as follows:
In above formula,For classification preference value real-time in the user of the i-th class article;User for the i-th class article
Real-time Brang Preference value;γ1、γ2、γ3、γ4And γ5For the corresponding weighted value of each real-time behavioral data.
Further, after the long-term preference of above-mentioned user and the real-time preference of user are calculated and completed, for the ease of
Line computation cluster is called, and can store the long-term preference of user and the real-time preference of user to ginseng by key assignments of user identifier
It examines in distributed caching cluster 209 shown in Fig. 2;Wherein, distributed caching cluster can include Redis clusters, can also wrap
Other clusters are included, such as can be Tair clusters etc., there is no special restriction on this for this example.
Fig. 5 is schematically shown a kind of to be recommended to correspond to according to preference long-term in the user and the real-time preference of user for user
Article method flow diagram.Refering to what is shown in Fig. 5, it is user's recommendation pair according to preference long-term in user and the real-time preference of user
The article answered can include step S510- steps S530.
Refering to what is shown in Fig. 5, in step S510, processing is preset to the user real-time preference progress third and obtains user's meaning
Figure.
In this example embodiment, to user, real-time preference carries out third and presets processing and obtain user view and can wrap
It includes:Configuration information entropy function, and predict that the user anticipates according to described information entropy function and the real-time classification preference of the user
Figure.For example:
First, above- mentioned information entropy function is configured;Wherein, which for example can be:
Wherein, H (x) for real-time classification preference a certain in the real-time preference of user (or certain
One real-time Brang Preference) x information content;p(xi) for real-time classification preference a certain in the real-time preference of user (or a certain real-time product
Board preference) x probability function;Further, can analyze the intention of user can pass through calculating as a chance event
The comentropy of user preference, it is possible to predict the intention of user:If comentropy is bigger, it is overstepping the bounds of propriety to represent user preference distribution
It dissipates, is only strolled with table, without specific purchase intention;Opposite, if comentropy is smaller, represents user preference and compare collection
In, illustrating user, this is carved with more specific purchase intention, at this moment should keypoint recommendation and the relevant article of user's purchase intention.Into
One step, this example embodiment weighs the true intention of user using comentropy, and comentropy can remove uncertainty institute
The information content for needing the measurement of information content namely unknown event that may contain, an event or a system, are exactly one
Stochastic variable, it has certain uncertainty, needs the information content that introducing removes uncertainty more, then comentropy is higher,
It is on the contrary then lower.
With continued reference to shown in Fig. 5, in step S520, according to preference long-term in the user, the real-time preference of user and
User view obtains user and integrates preference.
In this example embodiment, when obtaining long-term preference, the real-time preference of user and user view in above-mentioned user
Afterwards, can be that certain proportionality coefficient is configured in long-term preference, the real-time preference of user and user view in user, then according to each
From proportionality coefficient obtain the synthesis preference of user.For example, the proportionality coefficient that the medium-term and long-term preference for user is configured is 0.3, uses
It is 0.4 that the proportionality coefficient of family real-time preference configuration, which is 0.3, is the proportionality coefficient of user view configuration, then can each proportionality coefficient
Obtain the synthesis preference of user.
With continued reference to shown in Fig. 5, in step S530, preference is integrated according to the user and is recommended and the user for user
The corresponding article of synthesis preference.
In this example embodiment, after user's synthesis preference is obtained, can preference be integrated according to user and be pushed away for user
Recommend corresponding article.For example, the synthesis preference of user is jacket for women surplus cashmere grey, then can be carried out according to each keyword
Then search once arranges satisfactory article in order;Further, it can be placed comprising institute in the position most started
There is the article of keyword;Then it arranges down successively, so that user finds satisfactory article.
It is user's recommendation pair that Fig. 6, which schematically shows another kind according to preference long-term in the user and the real-time preference of user,
The method flow diagram for the article answered.Refering to what is shown in Fig. 6, recommended according to preference long-term in user and the real-time preference of user for user
Corresponding article can also include step S610- steps S630.
It is long in the user after normalized according to different type of recommendation refering to what is shown in Fig. 6, in step S610
Phase preference and the real-time preference allocation ratio coefficient of user.
In this example embodiment, being calculated for the ease of the whole preference to user can be to long in the user
Phase preference and the real-time preference of user are normalized.It, can be according to different recommendation classes after the completion of normalized
Type, for preference long-term in the user after normalized and the real-time preference allocation ratio coefficient of user.For example, work as type of recommendation
Then can be that larger proportionality coefficient is configured in the real-time preference of user during classification preference real-time for user.
With continued reference to shown in Fig. 6, in step S620, according to preference long-term in the user after the normalized and
The proportionality coefficient of the real-time preference of user calculates user's entirety preference.
With continued reference to shown in Fig. 6, in step S630, recommended according to user's entirety preference for the user corresponding
Article.
In this example embodiment, above-mentioned steps S620 and step S630 are further illustrated by.
For example, a certain recommendation scene needs to recommend 30 commodity, the weight of some classification accounts for all user preference classification power
The ratio of weight is 0.2, then the quantity of this classification interception commodity set is 6, if interception quantity is decimal, is taken upwards
It is whole;Finally commodity set is merged from big to small according to the weighted value of user's classification preference, by the result after merging by specific
The demand final process of business scenario is (for example:Fill, break up) after the commercial product recommending result that is needed;
In another example by taking classification is recommended as an example, according to current recommendation scene, corresponding level is found out from user preference classification
Classification, sort from big to small by weight, just obtain classification recommendation results, such as result deficiency, popular classification can be used to fill plan
Slightly;
For another example by taking activity recommendation as an example, the classification distribution vector for calculating each activity first (can be according to activity
Under commodity be mapped to classification weighted sum and obtain), by the preference classification of the classification distribution vector of each activity and user vector
A cartesian product is asked to obtain user to this activity preference weight, activity recommendation result is finally obtained according to weight sequencing.With
Family, which is intended to analysis, can be used as auxiliary that can be used for supplementing and intervening recommendation results, for example, identification active user at will strolls, it can
To recommend some random popular commodity to him, rich requirement is higher, conversely, identifying that user has specific purchase to incline
To, at this time can by recommendation results hit user tendency commodity put forward power, user is exposed to bigger preference for probability.
It should be understood that the term " first " used in the present invention, " second " only play the purpose of differentiation, it is not intended to
It and actual content is defined.
In conclusion the main advantage of the item recommendation method provided by the invention based on user preference is, on the one hand,
By the real-time preference for calculating user and long-term preference, real-time preference and long-term preference further according to user are recommended for user
Corresponding article preferably can recommend corresponding article for user according to the preference of user and demand, increase article and push away
Recommend the fineness of result so that user can select the article for meeting oneself demand within the shortest time, reduce user
The spending limit of user can also be increased simultaneously by selecting the time of article;On the other hand, user view is predicted, it can be with
The more accurate article needed for user recommended user improves the shopping experience of user;In another aspect, it is pushed away according to different
It recommends type and recommends corresponding article for user, personalized push away can be done to the article or brand and classification etc. of given set
It recommends, increases the interest of user's shopping.
Exemplary means
After the item recommendation method based on user preference of exemplary embodiment of the invention is described, next,
The article recommendation apparatus based on user preference of exemplary embodiment of the invention is described with reference to figure 7.
Refering to what is shown in Fig. 7, the article recommendation apparatus based on user preference of exemplary embodiment of the invention can include
710 Second processing module 720 of first processing module and article recommending module 730.Wherein:
First processing module 710 can be used for obtaining the first default processing of user's history behavioral data progress long in user
Phase preference.
Second processing module 720 can be used for the real-time behavioral data to user and carry out the second default processing to obtain user real-time
Preference.
Article recommending module 730 can be used for being pushed away for user according to preference long-term in the user and the real-time preference of user
Recommend corresponding article.
According to an exemplary embodiment of the present disclosure, article recommending module can be also used for:
Processing is preset to the user real-time preference progress third and obtains user view;
User is obtained according to preference long-term in the user, the real-time preference of user and user view and integrates preference;
Preference is integrated according to the user and recommends article corresponding with user synthesis preference for user.
According to an exemplary embodiment of the present disclosure, the behavioral data includes navigation patterns data, collection behavioral data, adds
It purchases a variety of in behavioral data, buying behavior data and exposure behavioral data.
According to an exemplary embodiment of the present disclosure, first processing module can be also used for:
Setup time attenuation function, and during according to the generations of the time attenuation function and each historical behavior data
Between and frequency calculate the first user preference value of each historical behavior data;
For each first weighted value of historical behavior data configuration, and according to each historical behavior data corresponding first
User preference value and the first weighted value calculate long-term preference in the user.
According to an exemplary embodiment of the present disclosure, Second processing module can be also used for:
It is calculated according to the time of origin and frequency of the time attenuation function and each real-time behavioral data
The second user preference value of each real-time behavioral data;
The second weighted value is configured, and according to each real-time behavioral data corresponding second for each real-time behavioral data
User preference value and the second weighted value calculate the real-time preference of user.
According to an exemplary embodiment of the present disclosure, the user preference includes user's classification preference and user brand is inclined
It is good;
Third is carried out to the real-time preference of the user and presets processing and obtain user view to include:
Configuration information entropy function, and the use is predicted according to described information entropy function and the real-time classification preference of the user
Family is intended to.
According to an exemplary embodiment of the present disclosure, the article recommendation apparatus based on user preference further includes:
Memory module can with user by preference long-term in the user, the real-time preference of user and user view store to point
In cloth cache cluster.
According to an exemplary embodiment of the present disclosure, the article recommendation apparatus based on user preference further includes:
Normalized module can be used for preference long-term in the user and the real-time preference of user is normalized
Processing.
According to an exemplary embodiment of the present disclosure, article recommending module can be also used for:
According to different type of recommendation, for preference long-term in the user after normalized and the real-time preference configuration of user
Proportionality coefficient;
It is calculated and used according to the proportionality coefficient of preference long-term in the user after the normalized and the real-time preference of user
Family entirety preference;
Corresponding article is recommended for the user according to user's entirety preference.
Each function module and above-mentioned side due to the article recommendation apparatus based on user preference of embodiment of the present invention
It is identical in method invention embodiment, therefore details are not described herein.
Exemplary storage medium
In the item recommendation method based on user preference for describing exemplary embodiment of the invention and based on user
After the article recommendation apparatus of preference, next, being said with reference to figure 8 to the storage medium of exemplary embodiment of the invention
It is bright.
Refering to what is shown in Fig. 8, describe the program product for being used to implement the above method according to the embodiment of the present invention
800, portable compact disc read only memory (CD-ROM) may be used and including program code, and can in terminal device,
Such as it is run on PC.However, the program product of the present invention is without being limited thereto.
The arbitrary combination of one or more readable mediums may be used in described program product.Readable medium can be readable letter
Number medium or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or
System, device or the device of semiconductor or arbitrary above combination.The more specific example of readable storage medium storing program for executing is (non exhaustive
List) include:It is electrical connection, portable disc, hard disk, random access memory (RAM) with one or more conducting wires, read-only
Memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory
(CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
Computer-readable signal media can include in a base band or as a carrier wave part propagation data-signal,
In carry readable program code.The data-signal of this propagation may be used diversified forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Readable signal medium can also be any readable Jie other than readable storage medium storing program for executing
Matter.
It can combine to write to perform the program that the present invention operates with the arbitrary of one or more programming languages
Code, described program design language include object oriented program language-Java, C++ etc., further include conventional
Procedural programming language-such as " C " language or similar programming language.Program code can be fully in user
It performs on computing device, partly perform on a user device, part performs or on a remote computing completely long-range
It is performed on computing device or server.In the situation for being related to remote computing device, remote computing device can be by any number of
The network of class including LAN (LAN) or wide area network (WAN), is connected to user calculating equipment.
Example electronic device
After the storage medium of exemplary embodiment of the invention is described, next, reference chart is to example of the present invention
The electronic equipment of property embodiment illustrates.
After the storage medium of exemplary embodiment of the invention is described, next, with reference to figure 9 to example of the present invention
The electronic equipment of property embodiment illustrates.
The electronic equipment 900 that Fig. 9 is shown is only an example, should not be to the function and use scope of the embodiment of the present invention
Bring any restrictions.
As shown in figure 9, electronic equipment 900 is showed in the form of universal computing device.The component of electronic equipment 900 can wrap
It includes but is not limited to:Above-mentioned at least one processing unit 910, above-mentioned at least one storage unit 920, connection different system component
The bus 930 of (including storage unit 920 and processing unit 910), display unit 940.
Wherein, the storage unit 920 has program stored therein code, and said program code can be by the processing unit 910
It performs so that the processing unit 910 performs each according to the present invention described in above-mentioned " illustrative methods " part of this specification
The step of kind illustrative embodiments.For example, the processing unit 910 can perform step S110- as shown in fig. 1
S130。
Storage unit 920 can include volatile memory cell, such as Random Access Storage Unit (RAM) 9201 and/or
Cache memory unit 9202 can further include read-only memory unit (ROM) 9203.
Storage unit 920 can also include program/utility with one group of (at least one) program module 9205
9204, such program module 9205 includes but not limited to:Operating system, one or more application program, other program moulds
Block and program data may include the realization of network environment in each or certain combination in these examples.
Bus 930 can include data/address bus, address bus and controlling bus.
Electronic equipment 900 can also be by input/output (I/O) interface 950, with one or more 800 (examples of external equipment
Such as keyboard, sensing equipment, bluetooth equipment) communication.Also, electronic equipment 1000 can also pass through network adapter 960 and one
A or multiple networks (such as LAN (LAN), wide area network (WAN) and/or public network, such as internet) communication.Such as figure
Shown, network adapter 960 is communicated by bus 930 with other modules of electronic equipment 900.Although it should be understood that in figure not
It shows, electronic equipment 900 can be combined and use other hardware and/or software module, including but not limited to:Microcode, equipment are driven
Dynamic device, redundant processing unit, external disk drive array, RAID system, tape drive and data backup storage system etc..
It should be noted that although being referred to several modules or submodule of pop-up processing unit in above-detailed,
Be it is this division be only exemplary it is not enforceable.In fact, according to the embodiment of the present invention, above-described two
The feature and function of a or more units/modules can embody in a units/modules.Conversely, above-described one
The feature and function of units/modules can be further divided into being embodied by multiple units/modules.
It should be noted that although being referred to several units/modules of device or subelement/module in above-detailed,
But it is this division be only exemplary it is not enforceable.In fact, according to the embodiment of the present invention, it is above-described
The feature and function of two or more units/modules can embody in a units/modules.Conversely, above-described one
The feature and function of a units/modules can be further divided into being embodied by multiple units/modules.
In addition, although the operation of the method for the present invention is described with particular order in the accompanying drawings, this do not require that or
The operation that these operations must be performed or have to carry out shown in whole according to the particular order by implying could be realized desired
As a result.Additionally or alternatively, it is convenient to omit multiple steps are merged into a step and performed and/or by one by certain steps
Step is decomposed into execution of multiple steps.
Although describe spirit and principles of the present invention by reference to several specific embodiments, it should be appreciated that, this
Invention is not limited to invented specific embodiment, does not also mean that the feature in these aspects cannot to the division of various aspects
Combination is this to divide merely to the convenience of statement to be benefited.The present invention is directed to cover appended claims spirit and
In the range of included various modifications and equivalent arrangements.
Claims (10)
1. a kind of item recommendation method based on user preference, including:
The first default processing is carried out to user's history behavioral data and obtains long-term preference in user;
To user, real-time behavioral data carries out the second default processing and obtains the real-time preference of user;
Corresponding article is recommended for user according to preference long-term in the user and the real-time preference of user.
2. the item recommendation method according to claim 1 based on user preference, wherein, according to long-term partially in the user
The good and real-time preference of user recommends corresponding article to include for user:
Processing is preset to the user real-time preference progress third and obtains user view;
User is obtained according to preference long-term in the user, the real-time preference of user and user view and integrates preference;
Preference is integrated according to the user and recommends article corresponding with user synthesis preference for user.
3. the item recommendation method according to claim 2 based on user preference, wherein, the behavioral data includes browsing
It is a variety of in behavioral data, collection behavioral data plus purchase behavioral data, buying behavior data and exposure behavioral data.
4. the item recommendation method according to claim 3 based on user preference, wherein, to the user's history behavior number
Long-term preference in user is obtained according to the default processing of progress first to include:
Setup time attenuation function, and according to the time attenuation function and the time of origin of each historical behavior data with
And frequency calculates the first user preference value of each historical behavior data;
For each first weighted value of historical behavior data configuration, and according to corresponding first user of each historical behavior data
Preference value and the first weighted value calculate long-term preference in the user.
5. the item recommendation method according to claim 4 based on user preference, wherein, the real-time behavior number to the user
The real-time preference of user is obtained according to the default processing of progress second to include:
Each institute is calculated according to the time of origin and frequency of the time attenuation function and each real-time behavioral data
State the second user preference value of real-time behavioral data;
The second weighted value is configured for each real-time behavioral data, and according to the corresponding second user of each real-time behavioral data
Preference value and the second weighted value calculate the real-time preference of user.
6. the item recommendation method according to claim 5 based on user preference, wherein, the user preference includes user
Classification preference and user brand preference;
Third is carried out to the real-time preference of the user and presets processing and obtain user view to include:
Configuration information entropy function, and predict that the user anticipates according to described information entropy function and the real-time classification preference of the user
Figure.
7. according to any item recommendation methods based on user preference of claim 4-6, wherein, the attenuation function
For:
Wherein, β is time constant;tuiGeneration time for user behavior data;T is the time for calculating preference.
8. a kind of article recommendation apparatus based on user preference, including:
First processing module obtains long-term preference in user for carrying out the first default processing to user's history behavioral data;
Second processing module carries out the second default processing for behavioral data real-time to user and obtains the real-time preference of user;
Article recommending module, for recommending corresponding object for user according to preference long-term in the user and the real-time preference of user
Product.
9. a kind of storage medium, is stored thereon with computer program, wherein, the computer program is realized when being executed by processor
The item recommendation method based on user preference described in any one of claim 1 to 7.
10. a kind of electronic equipment, including:
Processor;And
Memory, for storing the executable instruction of the processor;
Wherein, the processor is configured to come described in perform claim requirement any one of 1 to 7 via the execution executable instruction
The item recommendation method based on user preference.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711450258.1A CN108198019A (en) | 2017-12-27 | 2017-12-27 | Item recommendation method and device, storage medium, electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711450258.1A CN108198019A (en) | 2017-12-27 | 2017-12-27 | Item recommendation method and device, storage medium, electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108198019A true CN108198019A (en) | 2018-06-22 |
Family
ID=62584667
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711450258.1A Pending CN108198019A (en) | 2017-12-27 | 2017-12-27 | Item recommendation method and device, storage medium, electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108198019A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108898496A (en) * | 2018-07-05 | 2018-11-27 | 北京京东金融科技控股有限公司 | Products Show method, apparatus, medium and electronic equipment |
CN109064278A (en) * | 2018-07-26 | 2018-12-21 | 北京三快在线科技有限公司 | Target object recommended method and device, electronic equipment, storage medium |
CN109087162A (en) * | 2018-07-05 | 2018-12-25 | 杭州朗和科技有限公司 | Data processing method, system, medium and calculating equipment |
CN109359246A (en) * | 2018-12-07 | 2019-02-19 | 上海宏原信息科技有限公司 | A kind of brand cohesion calculation method based on forum user speech |
CN109543132A (en) * | 2018-11-22 | 2019-03-29 | 深圳墨世科技有限公司 | Content recommendation method, device, electronic equipment and storage medium |
CN110175895A (en) * | 2019-05-31 | 2019-08-27 | 京东方科技集团股份有限公司 | A kind of item recommendation method and device |
CN110210905A (en) * | 2019-05-31 | 2019-09-06 | 拉扎斯网络科技(上海)有限公司 | Characteristic similarity calculation method, device, electronic equipment and computer storage medium |
CN110222257A (en) * | 2019-05-13 | 2019-09-10 | 特斯联(北京)科技有限公司 | A kind of method, apparatus and data-link node for recommending business information |
CN111080339A (en) * | 2019-11-18 | 2020-04-28 | 口口相传(北京)网络技术有限公司 | Method and device for generating category preference data based on scene |
CN111582984A (en) * | 2020-05-08 | 2020-08-25 | 广州探途网络技术有限公司 | Commodity recommendation method and device and computing equipment |
CN111951049A (en) * | 2020-08-14 | 2020-11-17 | 中国工商银行股份有限公司 | Commodity similarity-based recommendation method and device |
CN111966903A (en) * | 2020-08-18 | 2020-11-20 | 中国银行股份有限公司 | Application software function recommendation method and device |
CN113546414A (en) * | 2021-07-30 | 2021-10-26 | 网易(杭州)网络有限公司 | Equipment purchasing method and device for game match, storage medium and computer equipment |
CN113806605A (en) * | 2021-09-23 | 2021-12-17 | 深圳供电局有限公司 | Content recommendation method and system based on digital historical information |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102368256A (en) * | 2010-09-30 | 2012-03-07 | 微软公司 | Providing suggestions based on intentions of subscribers |
WO2012118609A1 (en) * | 2011-03-01 | 2012-09-07 | Spectrum Bridge, Inc. | System and method for providing data to a portable communications device based on real-time subscriber behavior |
CN103678647A (en) * | 2013-12-20 | 2014-03-26 | Tcl集团股份有限公司 | Method and system for recommending information |
CN104123325A (en) * | 2013-04-28 | 2014-10-29 | 北京百度网讯科技有限公司 | Method for recommending multi-media files and recommendation server |
CN104679743A (en) * | 2013-11-26 | 2015-06-03 | 阿里巴巴集团控股有限公司 | Method and device for determining preference model of user |
CN105469263A (en) * | 2014-09-24 | 2016-04-06 | 阿里巴巴集团控股有限公司 | Commodity recommendation method and device |
CN106534902A (en) * | 2016-12-14 | 2017-03-22 | 北京数码视讯软件技术发展有限公司 | Behavior analysis method and system |
CN106850750A (en) * | 2016-12-26 | 2017-06-13 | 北京五八信息技术有限公司 | A kind of method and apparatus of real time propelling movement information |
CN107066476A (en) * | 2016-12-13 | 2017-08-18 | 江苏途致信息科技有限公司 | A kind of real-time recommendation method based on article similarity |
-
2017
- 2017-12-27 CN CN201711450258.1A patent/CN108198019A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102368256A (en) * | 2010-09-30 | 2012-03-07 | 微软公司 | Providing suggestions based on intentions of subscribers |
WO2012118609A1 (en) * | 2011-03-01 | 2012-09-07 | Spectrum Bridge, Inc. | System and method for providing data to a portable communications device based on real-time subscriber behavior |
CN104123325A (en) * | 2013-04-28 | 2014-10-29 | 北京百度网讯科技有限公司 | Method for recommending multi-media files and recommendation server |
CN104679743A (en) * | 2013-11-26 | 2015-06-03 | 阿里巴巴集团控股有限公司 | Method and device for determining preference model of user |
CN103678647A (en) * | 2013-12-20 | 2014-03-26 | Tcl集团股份有限公司 | Method and system for recommending information |
CN105469263A (en) * | 2014-09-24 | 2016-04-06 | 阿里巴巴集团控股有限公司 | Commodity recommendation method and device |
CN107066476A (en) * | 2016-12-13 | 2017-08-18 | 江苏途致信息科技有限公司 | A kind of real-time recommendation method based on article similarity |
CN106534902A (en) * | 2016-12-14 | 2017-03-22 | 北京数码视讯软件技术发展有限公司 | Behavior analysis method and system |
CN106850750A (en) * | 2016-12-26 | 2017-06-13 | 北京五八信息技术有限公司 | A kind of method and apparatus of real time propelling movement information |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108898496A (en) * | 2018-07-05 | 2018-11-27 | 北京京东金融科技控股有限公司 | Products Show method, apparatus, medium and electronic equipment |
CN109087162A (en) * | 2018-07-05 | 2018-12-25 | 杭州朗和科技有限公司 | Data processing method, system, medium and calculating equipment |
CN109064278A (en) * | 2018-07-26 | 2018-12-21 | 北京三快在线科技有限公司 | Target object recommended method and device, electronic equipment, storage medium |
CN109064278B (en) * | 2018-07-26 | 2021-04-02 | 北京三快在线科技有限公司 | Target object recommendation method and device, electronic equipment and storage medium |
CN109543132A (en) * | 2018-11-22 | 2019-03-29 | 深圳墨世科技有限公司 | Content recommendation method, device, electronic equipment and storage medium |
CN109359246A (en) * | 2018-12-07 | 2019-02-19 | 上海宏原信息科技有限公司 | A kind of brand cohesion calculation method based on forum user speech |
CN110222257B (en) * | 2019-05-13 | 2020-01-31 | 特斯联(北京)科技有限公司 | method and device for recommending service information and data link node |
CN110222257A (en) * | 2019-05-13 | 2019-09-10 | 特斯联(北京)科技有限公司 | A kind of method, apparatus and data-link node for recommending business information |
CN110210905A (en) * | 2019-05-31 | 2019-09-06 | 拉扎斯网络科技(上海)有限公司 | Characteristic similarity calculation method, device, electronic equipment and computer storage medium |
WO2020238502A1 (en) * | 2019-05-31 | 2020-12-03 | 京东方科技集团股份有限公司 | Article recommendation method and apparatus, electronic device and storage medium |
CN110175895A (en) * | 2019-05-31 | 2019-08-27 | 京东方科技集团股份有限公司 | A kind of item recommendation method and device |
CN111080339A (en) * | 2019-11-18 | 2020-04-28 | 口口相传(北京)网络技术有限公司 | Method and device for generating category preference data based on scene |
CN111080339B (en) * | 2019-11-18 | 2024-01-30 | 口口相传(北京)网络技术有限公司 | Scene-based category preference data generation method and device |
CN111582984A (en) * | 2020-05-08 | 2020-08-25 | 广州探途网络技术有限公司 | Commodity recommendation method and device and computing equipment |
CN111951049A (en) * | 2020-08-14 | 2020-11-17 | 中国工商银行股份有限公司 | Commodity similarity-based recommendation method and device |
CN111966903A (en) * | 2020-08-18 | 2020-11-20 | 中国银行股份有限公司 | Application software function recommendation method and device |
CN113546414A (en) * | 2021-07-30 | 2021-10-26 | 网易(杭州)网络有限公司 | Equipment purchasing method and device for game match, storage medium and computer equipment |
CN113806605A (en) * | 2021-09-23 | 2021-12-17 | 深圳供电局有限公司 | Content recommendation method and system based on digital historical information |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108198019A (en) | Item recommendation method and device, storage medium, electronic equipment | |
CN108629665B (en) | Personalized commodity recommendation method and system | |
US20220114199A1 (en) | System and method for information recommendation | |
US10475102B2 (en) | Providing personalized item recommendations using scalable matrix factorization with randomness | |
CN111461841B (en) | Article recommendation method, device, server and storage medium | |
CN110532479A (en) | A kind of information recommendation method, device and equipment | |
CN111080413A (en) | E-commerce platform commodity recommendation method and device, server and storage medium | |
CN108205768A (en) | Database building method and data recommendation method and device, equipment and storage medium | |
CN110008397B (en) | Recommendation model training method and device | |
CN109597973A (en) | A kind of recommendation, generation method and the device of official documents and correspondence information | |
CN112800342B (en) | Recommendation method, system, computer device and storage medium based on heterogeneous information | |
CN111737418B (en) | Method, apparatus and storage medium for predicting relevance of search term and commodity | |
US20190278819A1 (en) | Systems and methods of providing recommendations of content items | |
CN110348895A (en) | A kind of personalized recommendation method based on user tag, device and electronic equipment | |
CN107818491A (en) | Electronic installation, Products Show method and storage medium based on user's Internet data | |
KR102409718B1 (en) | Method and system for managing personal review shop on commerce platform connected with social network service | |
KR102409717B1 (en) | Method and apparatus for recommending selling item to seller of on-line shopping mall | |
CN111400613A (en) | Article recommendation method, device, medium and computer equipment | |
CN111967914A (en) | User portrait based recommendation method and device, computer equipment and storage medium | |
CN111754278A (en) | Article recommendation method and device, computer storage medium and electronic equipment | |
CN105335518A (en) | Method and device for generating user preference information | |
CN110599291A (en) | Intelligent e-commerce operation system and method for efficiently and accurately fusing multiple markets of merchant and buyer | |
CN113781149A (en) | Information recommendation method and device, computer-readable storage medium and electronic equipment | |
CN112036987B (en) | Method and device for determining recommended commodity | |
CN117057855A (en) | Data processing method and related device |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20191107 Address after: 310012 G building, 10 floor, A building, Paradise Software Park, 3 West Road, Hangzhou, Xihu District, Zhejiang Applicant after: Alibaba (China) Co.,Ltd. Address before: Hangzhou City, Zhejiang province 310051 Binjiang District River Street No. 1786 Jianghan Road Longsheng Building Room 803 Applicant before: NETEASE KOALA (HANGZHOU) TECH CO.,LTD. |
|
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180622 |