CN106372961A - Commodity recommendation method and device - Google Patents
Commodity recommendation method and device Download PDFInfo
- Publication number
- CN106372961A CN106372961A CN201610712391.9A CN201610712391A CN106372961A CN 106372961 A CN106372961 A CN 106372961A CN 201610712391 A CN201610712391 A CN 201610712391A CN 106372961 A CN106372961 A CN 106372961A
- Authority
- CN
- China
- Prior art keywords
- category
- commodity
- targeted customer
- belonging
- belong
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a commodity recommendation method and device, and belongs to the field of the network technology. The method comprises the following steps of on the basis of a plurality of first user amounts, determining a commodity recommendation list, wherein each first user amount is the amount of users who purchase first-category commodities and execute an appointed behavior for each second-category commodity in a plurality of second-category commodities in a preset time period, and the commodity recommendation list comprises N second-category marks; for each second category in N second categories, on the basis of the mark of the second category, determining the marks of the plurality of commodities which belong to the second category; on the basis of target user characteristic information and the commodity characteristic information of the plurality of commodities which belong to the second category, determining a plurality of recommended purchase probabilities through an appointed logistic regression model; and on the basis of the plurality of recommended purchase probabilities, recommending a target commodity in the plurality of commodities which belong to the second category to the target user. Therefore, different commodities are recommended to different users in a targeted way, and commodity recommendation efficiency is improved.
Description
Technical field
It relates to networking technology area, more particularly, to a kind of Method of Commodity Recommendation and device.
Background technology
With the fast development of network technology and e-commerce technology, Network is used widely.In network industry
In business, the merchandise news wanting to sell generally is issued by supplier on website, user can be believed based on this website and this commodity
Breath, buys this commodity in a network.However, at present, because different suppliers all can carry out Network on network, that is, not
Similar and different merchandise news can be issued on different websites with supplier, therefore so that occurring in that in Network
The data of magnanimity, so, leads to user in picking commodities, be easily trapped into vast and hazy in.
For this reason, in order to solve the above problems, that is, in Network, in order to improve purchase efficiency and the user of user
Experience, needs a solution badly, at present with the characteristic information of the history purchasing behavior according to historic user and user to be recommended
(for example, sex, age, income etc.), pointedly recommends possible commodity interested for user to be recommended, for example, for same
One commodity, are that the user of different sexes recommends different colours.
Content of the invention
For overcoming problem present in correlation technique, the disclosure provides a kind of Method of Commodity Recommendation and device.
In a first aspect, providing a kind of Method of Commodity Recommendation, methods described includes:
Based on multiple first user quantity, determine commercial product recommending list, each first user quantity is to buy first category
Commodity and in preset time period, specifies behavior is executed to the commodity of each second category in the commodity of multiple second category
Number of users, described commercial product recommending list includes the mark of n second category, and described n is more than or equal to 1 and is less than or equal to institute
State the quantity of multiple second category；
For each second category in described n second category, based on the mark of described second category, determine and belong to institute
State the mark of multiple commodity of second category；
Based on the product features information of targeted customer's characteristic information and the multiple commodity belonging to described second category, by referring to
Determine Logic Regression Models, determine multiple recommendation purchase probabilities, each recommends purchase probability is will to belong to each of described second category
After individual commercial product recommending is to targeted customer, described targeted customer buys the probability of each commodity belonging to described second category；
Based on the plurality of recommendation purchase probability, the end article in the multiple commodity belonging to described second category is recommended
To described targeted customer.
Alternatively, the feature of the described characteristic information based on targeted customer and the multiple commodity belonging to described second category is believed
Breath, by specifying Logic Regression Models, before determining multiple recommendation purchase probabilities, also includes:
For each second category in described n second category, obtained before current time to belonging to described second
Multiple commodity of classification execute the historic user characteristic information of described specifies behavior and belong to multiple commodity of described second category
Product features information；
Based on the historic user characteristic information that the multiple commodity belonging to described second category are executed with described specifies behaviors and
Belong to the product features information of multiple commodity of described second category, according to given combination strategy, generate multiple training characteristics to
Amount；
Based on the plurality of training feature vector, logic of propositions regression model is trained, obtains described specified logic
Regression model.
Alternatively, the feature of the described characteristic information based on targeted customer and the multiple commodity belonging to described second category is believed
Breath, by specifying Logic Regression Models, determines multiple recommendation purchase probabilities, comprising:
For each commodity in the multiple commodity belonging to described second category, the product features information based on described commodity
With described targeted customer's characteristic information, according to given combination strategy, generate target feature vector；
Based on described target feature vector, by described specified Logic Regression Models, determine that the recommendation of described commodity is bought
Probability.
Alternatively, described based on the plurality of recommendation purchase probability, by the multiple commodity belonging to described second category
End article recommends described targeted customer, including；
The maximum commodity of purchase probability are recommended to be defined as described target business in multiple commodity of described second category by belonging to
Product, described end article is recommended described targeted customer；Or,
Based on the plurality of commodity price recommending purchase probability with the multiple commodity belonging to described second category, determine institute
State multiple commodity income numerical value, each commodity is taken in numerical value and is used for instruction and will belong to each commodity of described second category to described
Targeted customer recommends the real revenue after preset times；Commodity income numerical value will be belonged in multiple commodity of described second category
Big commodity are defined as described end article, and described end article is recommended described targeted customer.
Alternatively, described based on multiple first user quantity, determine commercial product recommending list, comprising:
Based on multiple first user quantity, determine each Equations of The Second Kind in described first category and the plurality of second category
Similarity between not；
From the plurality of second category, the similarity and described first category between is selected to be more than or equal to predetermined threshold value
N second category；
Based on the mark of selected n second category, generate described commercial product recommending list.
Second aspect, provides a kind of device for recommending the commodity, and described device includes:
First determining module, for based on multiple first user quantity, determining commercial product recommending list, each first user number
Measure the business for buying each second category in the commodity of first category the commodity to multiple second category in preset time period
Product execute the number of users of specifies behavior, and described commercial product recommending list includes the mark of n second category, and described n is more than etc.
In 1 and less than or equal to the plurality of second category quantity；
Second determining module, for each Equations of The Second Kind in the n second category for described first determining module determination
Not, the mark based on described second category, determines the mark of the multiple commodity belonging to described second category；
3rd determining module, for the business based on targeted customer's characteristic information and the multiple commodity belonging to described second category
Product characteristic information, by specifying Logic Regression Models, determines multiple recommendation purchase probabilities, and each recommends purchase probability is to belong to
After each commercial product recommending of described second category is to targeted customer, described targeted customer buys and belongs to each of described second category
The probability of individual commodity；
Recommending module, for the multiple recommendation purchase probabilities determining based on described 3rd determining module, will belong to described the
End article in multiple commodity of two classifications recommends described targeted customer.
Alternatively, described device also includes:
Acquisition module, for for each second category in described n second category, it is right before current time to obtain
The multiple commodity belonging to described second category execute the historic user characteristic information of described specifies behavior and belong to described Equations of The Second Kind
The product features information of other multiple commodity；
Generation module, for being used based on the history that the multiple commodity belonging to described second category are executed with described specifies behavior
Family characteristic information and the product features information of the multiple commodity belonging to described second category, according to given combination strategy, generate many
Individual training feature vector；
Training module, for the plurality of training feature vector being generated based on described generation module, is returned to logic of propositions
Return model to be trained, obtain described specified Logic Regression Models.
Alternatively, described 3rd determining module includes:
First generation submodule, for for each commodity in the multiple commodity belonging to described second category, based on institute
State the product features information of commodity and described targeted customer's characteristic information, according to given combination strategy, generate target feature vector；
First determination sub-module, for based on described target feature vector, by described specified Logic Regression Models, determines
The recommendation purchase probability of described commodity.
Alternatively, described recommending module includes:
First recommendation submodule, for recommending the maximum business of purchase probability by belonging in multiple commodity of described second category
Product are defined as described end article, and described end article is recommended described targeted customer；Or,
Second recommendation submodule, for based on the plurality of multiple business recommending purchase probability and belonging to described second category
The commodity price of product, determines the plurality of commodity income numerical value, and each commodity income numerical value will belong to described second for instruction
Each commodity of classification recommend the real revenue after preset times to described targeted customer；The multiple of described second category will be belonged to
In commodity, the maximum commodity of commodity income numerical value are defined as described end article, and described end article is recommended described target
User.
Alternatively, described first determining module includes:
Second determination sub-module, for based on multiple first user quantity, determining described first category and the plurality of the
The similarity between each second category in two classifications；
Select submodule, big for from the plurality of second category, selecting the similarity and described first category between
In or be equal to predetermined threshold value n second category；
Second generation submodule, for the mark based on selected n second category, generates described commercial product recommending row
Table.
The third aspect, provides a kind of device for recommending the commodity, and described device includes:
Processor；
For storing the memorizer of processor executable；
Wherein, described processor is configured to:
Based on multiple first user quantity, determine commercial product recommending list, each first user quantity is to buy first category
Commodity and in preset time period, specifies behavior is executed to the commodity of each second category in the commodity of multiple second category
Number of users, described commercial product recommending list includes the mark of n second category, and described n is more than or equal to 1 and is less than or equal to institute
State the quantity of multiple second category；
For each second category in described n second category, based on the mark of described second category, determine and belong to institute
State the mark of multiple commodity of second category；
Based on the product features information of targeted customer's characteristic information and the multiple commodity belonging to described second category, by referring to
Determine Logic Regression Models, determine multiple recommendation purchase probabilities, each recommends purchase probability is will to belong to each of described second category
After individual commercial product recommending is to targeted customer, described targeted customer buys the probability of each commodity belonging to described second category；
Based on the plurality of recommendation purchase probability, the end article in the multiple commodity belonging to described second category is recommended
To described targeted customer.
The technical scheme that embodiment of the disclosure provides can include following beneficial effect:
In the disclosed embodiments, the commodity of the mark including n second category based on multiple first user quantity, are determined
Recommendation list, that is to say, predefines n second category from multiple second category, to be subsequently based on this n second category
Further determine that commodity to be recommended.I.e. for each second category in this n second category, the mark based on this second category
Know, determine the mark of the multiple commodity belonging to this second category, that is, this second category actually corresponds to multiple commodity, and the plurality of
The product features information of commodity typically each differs, the product characteristics letter based on targeted customer's characteristic information and the plurality of commodity
Breath, by having completed the specified Logic Regression Models trained in advance it may be determined that multiple recommendation purchase probability, due to each recommendation
Purchase probability be by belong to each commercial product recommending of this second category to targeted customer after, this targeted customer buys that to belong to this each
The probability of individual commodity, therefore, based on the plurality of recommendation purchase probability, can select end article from the plurality of commodity, and
This end article is recommended this targeted customer, that is to say, according to different user characteristic information, be subordinated to each second category
In multiple commodity, pointedly select different commercial product recommendings to different user, improve commercial product recommending efficiency.
It should be appreciated that above general description and detailed description hereinafter are only exemplary and explanatory, not
The disclosure can be limited.
Brief description
Accompanying drawing herein is merged in description and constitutes the part of this specification, shows the enforcement meeting the disclosure
Example, and be used for explaining the principle of the disclosure together with description.
Fig. 1 a is a kind of implementation environment schematic diagram according to an exemplary embodiment.
Fig. 1 b is a kind of flow chart of the Method of Commodity Recommendation according to an exemplary embodiment.
Fig. 2 a is a kind of flow chart of the Method of Commodity Recommendation implementing to exemplify according to another exemplary.
Fig. 2 b is a kind of schematic diagram of the training feature vector involved by Fig. 2 a embodiment.
Fig. 2 c is a kind of schematic diagram of the target feature vector involved by Fig. 2 a embodiment.
Fig. 3 a is a kind of block diagram of the device for recommending the commodity according to an exemplary embodiment.
Fig. 3 b is a kind of block diagram of the device for recommending the commodity implementing to exemplify according to another exemplary.
Fig. 4 is a kind of block diagram of the device for recommending the commodity 400 according to an exemplary embodiment.
Specific embodiment
Here will in detail exemplary embodiment be illustrated, its example is illustrated in the accompanying drawings.Explained below is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the disclosure.On the contrary, they be only with such as appended
The example of the consistent apparatus and method of some aspects being described in detail in claims, the disclosure.
Before the embodiment of the present disclosure is carried out with explanation explained in detail, first the application scenarios of the embodiment of the present disclosure are given
Explanation.
Fig. 1 a is a kind of implementation environment schematic diagram according to an exemplary embodiment.Mainly include in this implementation environment
User terminal 110 server 120, this user terminal 110 passes through between cable network or wireless network and this server 120
Set up communication connection.
Wherein, client or browser can have been run in this user terminal 110, permissible by this client or browser
Carry out web page access, thus realizing the purchase of commodity or the operation such as browsing.This user terminal 110 can be smart mobile phone, flat board
The equipment such as computer or computer, the embodiment of the present disclosure is not construed as limiting to this.
Wherein, this server 120 can be a server or the server set being made up of some servers
Group, this server 120 is mainly used in realizing the Method of Commodity Recommendation that the embodiment of the present disclosure is provided.
Fig. 1 b is a kind of flow chart of the Method of Commodity Recommendation according to an exemplary embodiment, as shown in Figure 1 b, should
Method of Commodity Recommendation is used in server, including following step:
In a step 101, based on multiple first user quantity, determine commercial product recommending list, each first user quantity is
Buy the commodity of first category and in preset time period, the commodity of each second category in the commodity of multiple second category are held
The number of users of row specifies behavior, this commercial product recommending list includes the mark of n second category, and this n is more than or equal to 1 and is less than
Or it is equal to the quantity of the plurality of second category.
In a step 102, for each second category in this n second category, based on the mark of this second category, really
Surely belong to the mark of multiple commodity of this second category.
In step 103, the product features based on targeted customer's characteristic information and the multiple commodity belonging to this second category
Information, by specify Logic Regression Models, determine multiple recommendation purchase probabilities, each recommend purchase probability be by belong to this second
After each commercial product recommending of classification is to targeted customer, this targeted customer buys the general of each commodity belonging to this second category
Rate.
At step 104, based on the plurality of recommendation purchase probability, by the target in the multiple commodity belonging to this second category
Commercial product recommending gives this targeted customer.
In the disclosed embodiments, the commodity of the mark including n second category based on multiple first user quantity, are determined
Recommendation list, that is to say, predefines n second category from multiple second category, to be subsequently based on this n second category
Further determine that commodity to be recommended.I.e. for each second category in this n second category, the mark based on this second category
Know, determine the mark of the multiple commodity belonging to this second category, that is, this second category actually corresponds to multiple commodity, and the plurality of
The product features information of commodity typically each differs, the product characteristics letter based on targeted customer's characteristic information and the plurality of commodity
Breath, by having completed the specified Logic Regression Models trained in advance it may be determined that multiple recommendation purchase probability, due to each recommendation
Purchase probability be by belong to each commercial product recommending of this second category to targeted customer after, this targeted customer buys that to belong to this each
The probability of individual commodity, therefore, based on the plurality of recommendation purchase probability, can select end article from the plurality of commodity, and
This end article is recommended this targeted customer, that is to say, according to different user characteristic information, be subordinated to each second category
In multiple commodity, pointedly select different commercial product recommendings to different user, improve commercial product recommending efficiency.
Alternatively, the characteristic information based on targeted customer and the characteristic information of the multiple commodity belonging to this second category, lead to
Cross and specify Logic Regression Models, before determining multiple recommendation purchase probabilities, also include:
For each second category in this n second category, obtain purchase before current time and belong to this Equations of The Second Kind
The historic user characteristic information of other multiple commodity and belong to this second category multiple commodity product features information；
Belong to the historic user characteristic information of multiple commodity of this second category based on purchase and belong to this second category
The product features information of multiple commodity, according to given combination strategy, generates multiple training feature vector；
Based on the plurality of training feature vector, logic of propositions regression model is trained, obtains this specified logistic regression
Model.
Alternatively, the characteristic information based on targeted customer and the characteristic information of the multiple commodity belonging to this second category, lead to
Cross and specify Logic Regression Models, determine multiple recommendation purchase probabilities, comprising:
For each commodity in the multiple commodity belonging to this second category, the product features information based on this commodity and should
Targeted customer's characteristic information, according to given combination strategy, generates target feature vector；
Based on this target feature vector, by this specified Logic Regression Models, determine the recommendation purchase probability of this commodity.
Alternatively, based on the plurality of recommendation purchase probability, by the end article in the multiple commodity belonging to this second category
Recommend this targeted customer, including；
The maximum commodity of purchase probability are recommended to be defined as this end article in multiple commodity of this second category by belonging to, will
This end article recommends this targeted customer；Or,
Based on the plurality of commodity price recommending purchase probability with the multiple commodity belonging to this second category, determine the plurality of
Numerical value taken in by commodity, and each commodity income numerical value is used for instruction to be pushed away each commodity belonging to this second category to this targeted customer
Recommend the real revenue after preset times；Determine belonging to commodity in multiple commodity of this second category and taking in the maximum commodity of numerical value
For this end article, and this end article is recommended this targeted customer.
Alternatively, based on multiple first user quantity, determine commercial product recommending list, comprising:
Based on multiple first user quantity, determine each second category in this first category and the plurality of second category it
Between similarity；
From the plurality of second category, the similarity and this first category between is selected to be more than or equal to the n of predetermined threshold value
Individual second category；
Based on the mark of selected n second category, generate this commercial product recommending list.
Above-mentioned all optional technical schemes, all can be according to the alternative embodiment arbitrarily combining to form the disclosure, and the disclosure is real
Apply example this is no longer repeated one by one.
Fig. 2 a is a kind of flow chart of the Method of Commodity Recommendation implementing to exemplify according to another exemplary, as shown in Figure 2 a,
This Method of Commodity Recommendation is used in server, and this Method of Commodity Recommendation comprises the following steps:
In step 201, based on multiple first user quantity, determine commercial product recommending list.
Wherein, each first user quantity is to buy the commodity of first category and to multiple Equations of The Second Kinds in preset time period
In other commodity, the commodity of each second category execute the number of users of specifies behavior.
Wherein, preset time period can be arranged it is also possible to be set by Server Default by user is self-defined according to the actual requirements
Put, the embodiment of the present disclosure does not limit to this.
Above-mentioned specifies behavior can also be by user's self-defined setting according to the actual requirements, and for example, this specifies behavior can be
Purchasing behavior or navigation patterns.
Here, the commodity of above-mentioned first category simply mean to a kind of classification of commodity, actually this kind of commodity are not entered
Row divides in detail, and for example, the commodity of this first category can be mobile phone, but here not the color of this mobile phone of special instructions,
The information such as price that is to say, that the commodity of this first category only refer to mobile phone, do not say be what color, what
The mobile phone of price.The commodity of above-mentioned second category are similar to the implication of the commodity of this first category, repeat no more here.
This commercial product recommending list above-mentioned includes the mark of n second category, and this n is more than or equal to 1 and less than or equal to this
The quantity of multiple second category.Wherein, the mark of this second category is used for one second category of unique mark, and for example, this second
The mark of classification can be product_id.That is to say, in this step, be only based on multiple first user quantity, determine
The commodity of n second category, for example, the commodity of this n second category include mobile phone shell, cellular phone power supplies and earphone.
It should be noted that, above-mentioned based on the plurality of first user quantity, determine that commercial product recommending list can be using working in coordination with
Realizing, it realizes process and may include that based on multiple first user quantity filter method, determines this first category and the plurality of second
The similarity between each second category in classification, from the plurality of second category, selects the phase between this first category
Like degree more than or equal to n second category of predetermined threshold value, based on the mark of selected n second category, generate this commodity
Recommendation list.
Wherein, this predetermined threshold value can by user according to the actual requirements self-defined setting it is also possible to by this Server Default
Setting, the embodiment of the present disclosure does not limit to this.
Above-mentioned based on multiple first user quantity, determine each Equations of The Second Kind in this first category and the plurality of second category
The process that implements of the similarity between not may include that for each second category in the plurality of second category, services
Device is based on the plurality of first user quantity, is determined similar between this second category and this first category by equation below (1)
Degree:
Wherein, this j represents the similarity between this second category and this first category, uijRepresent the plurality of first user number
Amount, this uiRepresent the number of users of the commodity buying first category, this ujRepresent the number of users of the commodity buying this second category
Amount.
It should be noted that above-mentioned based on multiple first user quantity, determine this first category and the plurality of second category
In each second category between the implementation method of similarity be only exemplary, in another embodiment, can also be passed through it
Determining this similarity, the embodiment of the present disclosure does not limit its mode to this.
In the disclosed embodiments, second category is bigger with the similarity of first category, illustrates to have purchased this first category
Commodity after, the first user that this second category is executed with specifies behavior is more, that is to say, interested in this second category the
One user is more, and therefore, server, from the plurality of second category, selects the similarity and this first category between to be more than or wait
In n second category of predetermined threshold value, and the mark based on this n second category, generate this commercial product recommending list.
In a kind of possible implementation, after server selects this n second category, can according to similarity from big to
Little order, the mark of this n second category is ranked up, and obtains this commercial product recommending list.
In step 202., for each second category in this n second category, based on the mark of this second category, really
Surely belong to the mark of multiple commodity of this second category.
That is to say, in each second category, all include multiple commodity, for example, if the commodity of this second category are mobile phone shell,
Then because this mobile phone shell includes multiple color, therefore, for the mobile phone shell of each color, the mark of all corresponding commodity, should
The mark of commodity can be designated as goods_id, and that is, above-mentioned product_id corresponds to multiple goods_id.
The mark of this second category is corresponding with the mark of the multiple commodity belonging to this second category to be stored in this server,
This server obtains the mark of the corresponding multiple commodity of mark of each second category.
In step 203, the product features based on targeted customer's characteristic information and the multiple commodity belonging to this second category
Information, by specifying Logic Regression Models, determines multiple recommendation purchase probabilities.
Wherein, each recommend purchase probability be by belong to each commercial product recommending of this second category to targeted customer after,
This targeted customer buys the probability of each commodity belonging to this second category.
Wherein, targeted customer's characteristic information can include sex, occupation, age, income etc., and the embodiment of the present disclosure is to this not
Limit, this targeted customer's characteristic information can be filled in register account number by this targeted customer after, sent out by user terminal
Give this server, and be saved in data base by this server.
Wherein, each product features information is used for describing the feature of each commodity, and this each product features information is all permissible
Including commodity color, commodity price, commodity size, commercial specification etc., additionally, with reference to Fig. 2 b, each product features information also may be used
To include material, style etc., the embodiment of the present disclosure does not also limit to this.
The above-mentioned product features information based on targeted customer's characteristic information and the multiple commodity belonging to this second category, passes through
Specified Logic Regression Models, determine that multiple processes of realizing recommending purchase probabilities include: for belonging to the multiple of this second category
Each commodity in commodity, the product features information based on this commodity and this targeted customer's characteristic information, according to given combination plan
Slightly, generate target feature vector, based on this target feature vector, by this specified Logic Regression Models, determine pushing away of this commodity
Recommend purchase probability.
Wherein, the above-mentioned product features information based on this commodity and this targeted customer's characteristic information, according to given combination plan
Slightly, the process of realizing generating target feature vector may include that and refer to Fig. 2 c, in a kind of possible implementation, permissible
According to the actual requirements, the sex in the commodity color in this product features information and this targeted customer's characteristic information is carried out feature
Combination, and the income in the commodity price in this product features information and this targeted customer's characteristic information is carried out feature group
Close, obtain this target feature vector.
By combinations thereof mode, sex and the income of this targeted customer can be directed to, may recommend for this targeted customer
Color interested and the commodity of price.Its recommendation process includes: based on this target feature vector, by this specified logistic regression
Model it may be determined that each commercial product recommending of the features such as different colours, different price is given after this targeted customer, this targeted customer
It is likely to purchase the probability of each commodity.
It should be noted that above-mentioned by the commodity color in this product features information and this targeted customer's characteristic information
Sex carries out combinations of features, and by the income in the commodity price in this product features information and this targeted customer's characteristic information
It is only exemplary for carrying out combinations of features, and in another embodiment, this given combination strategy can also include other compound modes, this
Open embodiment does not limit to this.
In addition, it is necessary to explanation, above-mentioned based on this target feature vector, by this specified Logic Regression Models, determine
The process that implements of the recommendation purchase probability of this commodity may refer to correlation technique, and the embodiment of the present disclosure is not made to this in detail yet
Description.
Further, in addition it is also necessary to determine this specified Logic Regression Models before determining the plurality of recommendation purchase probability, its
In, determine that the process of realizing of this specified Logistic Regression module includes:
For each second category in this n second category, obtained before current time to belonging to this second category
Multiple commodity execute specifies behaviors historic user characteristic information and belong to this second category multiple commodity product features
Information, based on the historic user characteristic information that the multiple commodity belonging to this second category are executed with specifies behaviors and belong to this second
The product features information of multiple commodity of classification, according to given combination strategy, generates multiple training feature vector, based on the plurality of
Training feature vector, is trained to logic of propositions regression model, obtains this specified Logic Regression Models.
Wherein, above-mentioned given combination strategy can be arranged in advance in the server.
That is to say, for each second category in this n second category, due to before current time, there being some to go through
History user have purchased the multiple commodity belonging to this second category, or, also have some historic user be only browsed belong to this
Multiple commodity of two classifications, therefore, server can execute nominated bank based on all to each commodity belonging to this second category
For historic user characteristic information and each commodity product features information, generate training feature vector.
Wherein, during generating training feature vector it is also possible to according to the actual requirements, according to given combination strategy,
This historic user characteristic information and this product features information are carried out combinations of features, for example, refer to Fig. 2 b, this Fig. 2 b shows
One training feature vector, that is to say, can be by the color in the sex in historic user characteristic information and product features information
Carry out combinations of features, and the income in historic user characteristic information and price are carried out combinations of features, by this kind of combination side
Formula, the specified Logic Regression Models after training are it may be determined that the user of different sexes is to the commodity belonging to second category
Color preference, and there is the price preference to the commodity belonging to this second category for the different users taking in.
In addition, in the training feature vector being generated, in order to determine that how many historic user have purchased this commodity, also needs
The specifies behavior of historic user is marked, for example, in figure 2b, if historic user have purchased this commodity, server
In this purchase field correspondence position labelling " yes ", if this historic user has browsed this commodity, but do not buy this commodity, then service
Device is in this purchase field correspondence position labelling " no ".
It should be noted that during generating training feature vector, except being related to this historic user characteristic information and business
It is also possible to be related to contextual feature information and further feature information outside product characteristic information, the embodiment of the present disclosure does not limit to this
Fixed.
Wherein, based on the plurality of training feature vector, logic of propositions regression model is trained to realize process permissible
Including processes such as training feature vector conversion, logistic regression, generation models, in addition, obtaining specifying Logic Regression Models in training
Afterwards, this specified Logic Regression Models can also be estimated and the operation such as store, the embodiment of the present disclosure is not detailed Jie to this
Continue.
In step 204, based on the plurality of recommendation purchase probability, by the target in the multiple commodity belonging to this second category
Commercial product recommending gives this targeted customer.
During actual realization, for commodity provider, it may be desirable to reach difference during commercial product recommending
Recommendation purpose, for example, in a kind of possible implementation, carry out commercial product recommending purpose be improve commodity sale
Amount, and in alternatively possible implementation, the purpose carrying out commercial product recommending is to improve total sales volume, therefore, according to commodity
The purpose recommended is different, above-mentioned based on the plurality of recommendation purchase probability, by the target in the multiple commodity belonging to this second category
Commercial product recommending gives this targeted customer can be to include any one being implemented as described below in mode:
First kind of way: recommend the maximum commodity of purchase probability to be defined as this in multiple commodity of this second category by belonging to
End article, this end article is recommended this targeted customer.
That is to say, in this kind of implementation, due to recommending purchase probability bigger, illustrate to correspond to this recommendation purchase probability
Commercial product recommending give this targeted customer after, this targeted customer buy probability bigger, therefore, it can be appreciated that to improve business
The sales volume of product, then for each second category in multiple second category, can belong to multiple commodity of this second category
The middle commercial product recommending recommending purchase probability maximum, to targeted customer, so, can improve the sales volume of commodity.
The second way: based on the plurality of commodity valency recommending purchase probability with the multiple commodity belonging to this second category
Lattice, determine the plurality of commodity income numerical value, and each commodity income numerical value is used for instruction and will belong to each commodity of this second category
Recommend the real revenue after preset times to this targeted customer, commodity income numerical value in multiple commodity of this second category will be belonged to
Maximum commodity are defined as this end article, and this end article is recommended this targeted customer.
Wherein, this preset times can by user according to the actual requirements self-defined setting it is also possible to by this Server Default
Setting, the embodiment of the present disclosure is not construed as limiting to this.
From unlike the first implementation, in this kind of implementation, its main purpose is to improve total sale
Volume, that is to say, in the first implementation above-mentioned, although commodity income numerical value in multiple commodity of this second category will be belonged to
Maximum commodity are defined as this end article, but due to determined by the price of end article may be very low, so will be determined
End article recommend this targeted customer after, possible total sales volume is not maximum.
For this reason, in this kind of implementation, server based on the plurality of recommendation purchase probability and belongs to this second category
The commodity price of multiple commodity, determines the plurality of commodity income numerical value, it implements process and includes: each is recommended by server
Purchase probability is multiplied with the price of each commodity, and afterwards, by being multiplied, each product obtaining is multiplied with above-mentioned preset times, obtains
Belong to the commodity income numerical value of multiple commodity of this second category.
For example, in above-mentioned implementation, if the commodity price belonging to certain commodity of this second category is p, this is preset
Number of times is 1000 times, and the recommendation purchase probability of this commodity is b, then the commodity income numerical value of this commodity is r=b*p*1000, its
In, " * " multiplication should be represented, so, obtain the commodity income numerical value of this commodity.
If this commodity income numerical value is bigger, illustrate this commercial product recommending is bigger to the real revenue obtaining after targeted customer,
That is to say, the total sales volume obtaining is bigger, therefore, for each second category in multiple second category, server will belong to
In multiple commodity of this second category, the maximum commodity of commodity income numerical value are defined as this end article, so, not only can protect
Demonstrate,prove the commodity that recommended commodity are interested to this targeted customer, meanwhile, also improve total sales volume, that is, ensure that total sale
Volume maximizes.
In a kind of possible implementation, it is each end article belonging to each second category that targeted customer recommends
Commodity income numerical value can exist in the form of a list, for example, as shown in table 1 below.
Table 1
| user_id | Numerical value taken in by commodity |
| user_1 | goods_i:ri,goods_g:rg,... |
| ... | ... |
Wherein, user_1 represents targeted customer, and goods_i:ri represents the end article goods_i belonging to second category i
Commodity income numerical value be ri, in the same manner, this goods_g:rg represent belong to second category g end article goods_g commodity
Income numerical value is rg.
It is targeted customer's Recommendations above by different modes, different effects can be reached, increased commercial product recommending
Mode.
In the disclosed embodiments, the commodity of the mark including n second category based on multiple first user quantity, are determined
Recommendation list, that is to say, predefines n second category from multiple second category, to be subsequently based on this n second category
Further determine that commodity to be recommended.I.e. for each second category in this n second category, the mark based on this second category
Know, determine the mark of the multiple commodity belonging to this second category, that is, this second category actually corresponds to multiple commodity, and the plurality of
The product features information of commodity typically each differs, the product characteristics letter based on targeted customer's characteristic information and the plurality of commodity
Breath, by having completed the specified Logic Regression Models trained in advance it may be determined that multiple recommendation purchase probability, due to each recommendation
Purchase probability be by belong to each commercial product recommending of this second category to targeted customer after, this targeted customer buys that to belong to this each
The probability of individual commodity, therefore, based on the plurality of recommendation purchase probability, can select end article from the plurality of commodity, and
This end article is recommended this targeted customer, that is to say, according to different user characteristic information, be subordinated to each second category
In multiple commodity, pointedly select different commercial product recommendings to different user, improve commercial product recommending efficiency.
Fig. 3 a is a kind of block diagram of the device for recommending the commodity according to an exemplary embodiment.Reference picture 3a, this device
Including the first determining module 310, the second determining module 320, the 3rd determining module 330 and recommending module 340.
First determining module 310, for based on multiple first user quantity, determining commercial product recommending list, each first use
Amount amount is to buy the commodity of first category and to each second category in the commodity of multiple second category in preset time period
Commodity execute the number of users of specifies behavior, this commercial product recommending list includes the mark of n second category, this n more than etc.
In 1 and less than or equal to the plurality of second category quantity；
Second determining module 320, for each Equations of The Second Kind in n second category determining for this first determining module
Not, the mark based on this second category, determines the mark of the multiple commodity belonging to this second category；
3rd determining module 330, for based on targeted customer's characteristic information and the multiple commodity belonging to this second category
Product features information, by specifying Logic Regression Models, determines multiple recommendation purchase probabilities, and each recommends purchase probability is to belong to
After each commercial product recommending of this second category is to targeted customer, this targeted customer buys each business belonging to this second category
The probability of product；
Recommending module 340, for based on the 3rd determining module determine multiple recommendation purchase probabilities, will belong to this second
End article in multiple commodity of classification recommends this targeted customer.
Alternatively, refer to Fig. 3 b, this device also includes:
Acquisition module 350, for for each second category in this n second category, obtaining before current time
The multiple commodity belonging to this second category are executed with the historic user characteristic information of this specifies behavior and belongs to this second category
The product features information of multiple commodity；
Generation module 360, for being used based on the history that the multiple commodity belonging to this second category are executed with this specifies behavior
Family characteristic information and the product features information of the multiple commodity belonging to this second category, according to given combination strategy, generate multiple
Training feature vector；
Training module 370, for the plurality of training feature vector being generated based on this generation module, is returned to logic of propositions
Model is trained, and obtains this specified Logic Regression Models.
Alternatively, the 3rd determining module 330 includes:
First generation submodule, for for each commodity in the multiple commodity belonging to this second category, based on this business
The product features information of product and this targeted customer's characteristic information, according to given combination strategy, generate target feature vector；
First determination sub-module, for based on this target feature vector, by this specified Logic Regression Models, determining this business
The recommendation purchase probability of product.
Alternatively, this recommending module 340 includes:
First recommendation submodule, for recommending the maximum commodity of purchase probability by belonging in multiple commodity of this second category
It is defined as this end article, this end article is recommended this targeted customer；Or,
Second recommendation submodule, for recommending purchase probability and the multiple commodity belonging to this second category based on the plurality of
Commodity price, determines the plurality of commodity income numerical value, and each commodity income numerical value is used for instruction and will belong to each of this second category
Individual commodity recommend the real revenue after preset times to this targeted customer；Receive belonging to commodity in multiple commodity of this second category
Enter the maximum commodity of numerical value and be defined as this end article, and this end article is recommended this targeted customer.
Alternatively, this first determining module 310 includes:
Second determination sub-module, for based on multiple first user quantity, determining this first category and the plurality of Equations of The Second Kind
The similarity between each second category in not；
Select submodule, for from the plurality of second category, select the similarity and this first category between to be more than or
The n second category equal to predetermined threshold value；
Second generation submodule, for the mark based on selected n second category, generates this commercial product recommending list.
In the disclosed embodiments, the commodity of the mark including n second category based on multiple first user quantity, are determined
Recommendation list, that is to say, predefines n second category from multiple second category, to be subsequently based on this n second category
Further determine that commodity to be recommended.I.e. for each second category in this n second category, the mark based on this second category
Know, determine the mark of the multiple commodity belonging to this second category, that is, this second category actually corresponds to multiple commodity, and the plurality of
The product features information of commodity typically each differs, the product characteristics letter based on targeted customer's characteristic information and the plurality of commodity
Breath, by having completed the specified Logic Regression Models trained in advance it may be determined that multiple recommendation purchase probability, due to each recommendation
Purchase probability be by belong to each commercial product recommending of this second category to targeted customer after, this targeted customer buys that to belong to this each
The probability of individual commodity, therefore, based on the plurality of recommendation purchase probability, can select end article from the plurality of commodity, and
This end article is recommended this targeted customer, that is to say, according to different user characteristic information, be subordinated to each second category
In multiple commodity, pointedly select different commercial product recommendings to different user, improve commercial product recommending efficiency.
With regard to the device in above-described embodiment, wherein the concrete mode of modules execution operation is in relevant the method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
Fig. 4 is a kind of block diagram of the device for recommending the commodity 400 according to an exemplary embodiment.For example, device 400 can
To be provided as a server.With reference to Fig. 4, device 400 includes process assembly 422, and it further includes one or more process
Device, and the memory resource representated by memorizer 432, can be by the instruction of the execution of process assembly 422, such as storage
Application program.In memorizer 432 storage application program can include one or more each refer to corresponding to one group
The module of order.Additionally, process assembly 422 is configured to execute instruction, to execute involved by said method Fig. 1 b or Fig. 2 embodiment
Method of Commodity Recommendation.
Device 400 can also include the power management that a power supply module 426 is configured to performs device 400, and one has
Line or radio network interface 450 are configured to for device 400 to be connected to network, and input and output (i/o) interface 458.Dress
Put 400 can operate based on the operating system being stored in memorizer 432, such as windows servertm, mac os xtm,
unixtm,linuxtm, freebsdtmOr it is similar.
Those skilled in the art, after considering description and putting into practice invention disclosed herein, will readily occur to its of the disclosure
Its embodiment.The application is intended to any modification, purposes or the adaptations of the disclosure, these modifications, purposes or
Person's adaptations are followed the general principle of the disclosure and are included the undocumented common knowledge in the art of the disclosure
Or conventional techniques.Description and embodiments be considered only as exemplary, the true scope of the disclosure and spirit by following
Claim is pointed out.
It should be appreciated that the disclosure is not limited to be described above and precision architecture illustrated in the accompanying drawings, and
And various modifications and changes can carried out without departing from the scope.The scope of the present disclosure only to be limited by appended claim.
Claims (11)
1. a kind of Method of Commodity Recommendation is it is characterised in that methods described includes:
Based on multiple first user quantity, determine commercial product recommending list, each first user quantity is to buy the business of first category
Product simultaneously execute the user of specifies behavior to the commodity of each second category in the commodity of multiple second category in preset time period
Quantity, described commercial product recommending list includes the mark of n second category, and described n is more than or equal to 1 and less than or equal to described many
The quantity of individual second category；
For each second category in described n second category, based on the mark of described second category, determine and belong to described the
The mark of multiple commodity of two classifications；
Based on the product features information of targeted customer's characteristic information and the multiple commodity belonging to described second category, patrolled by specifying
Collect regression model, determine multiple recommendation purchase probabilities, each recommends purchase probability is to belong to each business of described second category
After product recommend targeted customer, described targeted customer buys the probability of each commodity belonging to described second category；
Based on the plurality of recommendation purchase probability, the end article in the multiple commodity belonging to described second category is recommended institute
State targeted customer.
2. the method for claim 1 is it is characterised in that the described characteristic information based on targeted customer and belong to described the
The characteristic information of multiple commodity of two classifications, by specifying Logic Regression Models, before determining multiple recommendation purchase probabilities, also wraps
Include:
For each second category in described n second category, obtained before current time to belonging to described second category
Multiple commodity execute described specifies behaviors historic user characteristic information and belong to described second category multiple commodity business
Product characteristic information；
Based on the historic user characteristic information of described specifies behaviors being executed to the multiple commodity belonging to described second category and belongs to
The product features information of multiple commodity of described second category, according to given combination strategy, generates multiple training feature vector；
Based on the plurality of training feature vector, logic of propositions regression model is trained, obtains described specified logistic regression
Model.
3. method as claimed in claim 1 or 2 is it is characterised in that the described characteristic information based on targeted customer and belong to institute
State the characteristic information of multiple commodity of second category, by specifying Logic Regression Models, determine multiple recommendation purchase probabilities, bag
Include:
For each commodity in the multiple commodity belonging to described second category, the product features information based on described commodity and institute
State targeted customer's characteristic information, according to given combination strategy, generate target feature vector；
Based on described target feature vector, by described specified Logic Regression Models, determine the recommendation purchase probability of described commodity.
4. the method for claim 1 it is characterised in that described based on the plurality of recommendation purchase probability, institute will be belonged to
The end article stated in multiple commodity of second category recommends described targeted customer, including；
The maximum commodity of purchase probability are recommended to be defined as described end article in multiple commodity of described second category by belonging to, will
Described end article recommends described targeted customer；Or,
Based on the plurality of commodity price recommending purchase probability with the multiple commodity belonging to described second category, determine described many
Numerical value taken in by individual commodity, and each commodity income numerical value is used for instruction and will belong to each commodity of described second category to described target
User recommends the real revenue after preset times；By belonging to commodity in multiple commodity of described second category, to take in numerical value maximum
Commodity are defined as described end article, and described end article is recommended described targeted customer.
5. the method for claim 1 it is characterised in that described based on multiple first user quantity, determine commercial product recommending
List, comprising:
Based on multiple first user quantity, determine each second category in described first category and the plurality of second category it
Between similarity；
From the plurality of second category, the similarity and described first category between is selected to be more than or equal to the n of predetermined threshold value
Individual second category；
Based on the mark of selected n second category, generate described commercial product recommending list.
6. a kind of device for recommending the commodity is it is characterised in that described device includes:
First determining module, for based on multiple first user quantity, determining commercial product recommending list, each first user quantity is
Buy the commodity of first category and in preset time period, the commodity of each second category in the commodity of multiple second category are held
The number of users of row specifies behavior, described commercial product recommending list includes the mark of n second category, described n be more than or equal to 1 and
Quantity less than or equal to the plurality of second category；
Second determining module, for each second category in the n second category for described first determining module determination, base
In the mark of described second category, determine the mark of the multiple commodity belonging to described second category；
3rd determining module, special for the commodity based on targeted customer's characteristic information and the multiple commodity belonging to described second category
Reference ceases, and by specifying Logic Regression Models, determines multiple recommendation purchase probabilities, and each recommends purchase probability is will to belong to described
After each commercial product recommending of second category is to targeted customer, described targeted customer buys each business belonging to described second category
The probability of product；
Recommending module, for the multiple recommendation purchase probabilities determining based on described 3rd determining module, will belong to described Equations of The Second Kind
End article in other multiple commodity recommends described targeted customer.
7. device as claimed in claim 6 is it is characterised in that described device also includes:
Acquisition module, for for each second category in described n second category, obtaining before current time to belonging to
Multiple commodity of described second category execute the historic user characteristic information of described specifies behavior and belong to described second category
The product features information of multiple commodity；
Generation module, for special based on the historic user that the multiple commodity belonging to described second category are executed with described specifies behavior
Reference cease and belong to described second category multiple commodity product features information, according to given combination strategy, generate multiple instructions
Practice characteristic vector；
Training module, for the plurality of training feature vector generating based on described generation module, returns mould to logic of propositions
Type is trained, and obtains described specified Logic Regression Models.
8. device as claimed in claims 6 or 7 is it is characterised in that described 3rd determining module includes:
First generation submodule, for for each commodity in the multiple commodity belonging to described second category, based on described business
The product features information of product and described targeted customer's characteristic information, according to given combination strategy, generate target feature vector；
First determination sub-module, for based on described target feature vector, by described specified Logic Regression Models, determines described
The recommendation purchase probability of commodity.
9. device as claimed in claim 6 is it is characterised in that described recommending module includes:
First recommendation submodule, for recommending the maximum commodity of purchase probability true by belonging in multiple commodity of described second category
It is set to described end article, described end article is recommended described targeted customer；Or,
Second recommendation submodule, for recommending purchase probability and the multiple commodity belonging to described second category based on the plurality of
Commodity price, determines the plurality of commodity income numerical value, and each commodity income numerical value will belong to described second category for instruction
Each commodity to described targeted customer recommend preset times after real revenue；Multiple commodity of described second category will be belonged to
Middle commodity are taken in the maximum commodity of numerical value and are defined as described end article, and described end article is recommended described target use
Family.
10. device as claimed in claim 6 is it is characterised in that described first determining module includes:
Second determination sub-module, for based on multiple first user quantity, determining described first category and the plurality of Equations of The Second Kind
The similarity between each second category in not；
Select submodule, for from the plurality of second category, select the similarity and described first category between to be more than or
The n second category equal to predetermined threshold value；
Second generation submodule, for the mark based on selected n second category, generates described commercial product recommending list.
A kind of 11. devices for recommending the commodity are it is characterised in that described device includes:
Processor；
For storing the memorizer of processor executable；
Wherein, described processor is configured to:
Based on multiple first user quantity, determine commercial product recommending list, each first user quantity is to buy the business of first category
Product simultaneously execute the user of specifies behavior to the commodity of each second category in the commodity of multiple second category in preset time period
Quantity, described commercial product recommending list includes the mark of n second category, and described n is more than or equal to 1 and less than or equal to described many
The quantity of individual second category；
For each second category in described n second category, based on the mark of described second category, determine and belong to described the
The mark of multiple commodity of two classifications；
Based on the product features information of targeted customer's characteristic information and the multiple commodity belonging to described second category, patrolled by specifying
Collect regression model, determine multiple recommendation purchase probabilities, each recommends purchase probability is to belong to each business of described second category
After product recommend targeted customer, described targeted customer buys the probability of each commodity belonging to described second category；
Based on the plurality of recommendation purchase probability, the end article in the multiple commodity belonging to described second category is recommended institute
State targeted customer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610712391.9A CN106372961A (en) | 2016-08-23 | 2016-08-23 | Commodity recommendation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610712391.9A CN106372961A (en) | 2016-08-23 | 2016-08-23 | Commodity recommendation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106372961A true CN106372961A (en) | 2017-02-01 |
Family
ID=57878937
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610712391.9A Pending CN106372961A (en) | 2016-08-23 | 2016-08-23 | Commodity recommendation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106372961A (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107341272A (en) * | 2017-08-25 | 2017-11-10 | 北京奇艺世纪科技有限公司 | A kind of method for pushing, device and electronic equipment |
| CN108053305A (en) * | 2018-02-12 | 2018-05-18 | 北京小度信息科技有限公司 | The data processing method and device promoted applied to trade company |
| CN108090807A (en) * | 2017-12-13 | 2018-05-29 | 北京小度信息科技有限公司 | Information recommendation method and device |
| CN108154420A (en) * | 2017-12-26 | 2018-06-12 | 泰康保险集团股份有限公司 | Products Show method and device, storage medium, electronic equipment |
| CN108198051A (en) * | 2018-03-01 | 2018-06-22 | 口碑(上海)信息技术有限公司 | Across the Method of Commodity Recommendation and device of merchandise classification |
| CN108304115A (en) * | 2018-02-12 | 2018-07-20 | 广东欧珀移动通信有限公司 | Terminal entry selection method, device and storage medium and terminal |
| CN108628863A (en) * | 2017-03-15 | 2018-10-09 | 百度在线网络技术（北京）有限公司 | Information acquisition method and device |
| CN108932530A (en) * | 2018-06-29 | 2018-12-04 | 新华三大数据技术有限公司 | The construction method and device of label system |
| CN108959340A (en) * | 2018-03-30 | 2018-12-07 | 深圳春沐源控股有限公司 | Page display method and page display system |
| CN109242649A (en) * | 2018-10-31 | 2019-01-18 | 广州品唯软件有限公司 | A kind of Method of Commodity Recommendation and relevant apparatus |
| CN109635153A (en) * | 2018-11-22 | 2019-04-16 | 腾讯音乐娱乐科技（深圳）有限公司 | Migration path generation method, device and storage medium |
| CN110069540A (en) * | 2017-09-08 | 2019-07-30 | 北京京东尚科信息技术有限公司 | Information output method and device |
| CN110111186A (en) * | 2019-05-10 | 2019-08-09 | 拉扎斯网络科技（上海）有限公司 | Information processing method, device, electronic equipment and computer readable storage medium |
| CN110134469A (en) * | 2018-02-09 | 2019-08-16 | 优视科技有限公司 | Switching method, device and the computer equipment of holiday theme |
| CN110519318A (en) * | 2018-05-22 | 2019-11-29 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN110838043A (en) * | 2019-11-05 | 2020-02-25 | 智者四海（北京）技术有限公司 | Commodity recommendation method and device |
| CN111009299A (en) * | 2019-12-17 | 2020-04-14 | 武汉海云健康科技股份有限公司 | Similar medicine recommendation method and system, server and medium |
| CN111199459A (en) * | 2019-12-30 | 2020-05-26 | 深圳市盟天科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
| CN111695922A (en) * | 2019-03-15 | 2020-09-22 | 北京沃东天骏信息技术有限公司 | Potential user determination method and device, storage medium and electronic equipment |
| CN111861623A (en) * | 2019-12-30 | 2020-10-30 | 北京骑胜科技有限公司 | Information recommendation method, device and equipment |
| US10922737B2 (en) | 2017-12-22 | 2021-02-16 | Industrial Technology Research Institute | Interactive product recommendation method and non-transitory computer-readable medium |
| CN113379516A (en) * | 2021-08-12 | 2021-09-10 | 永正信息技术(南京)有限公司 | Recommended product determination method and device |
| CN113610608A (en) * | 2021-08-19 | 2021-11-05 | 创优数字科技(广东)有限公司 | User preference recommendation method and device, electronic equipment and storage medium |
| CN115471302A (en) * | 2022-11-14 | 2022-12-13 | 山东智豆数字科技有限公司 | Electronic marketing data processing method based on big data analysis |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8452797B1 (en) * | 2011-03-09 | 2013-05-28 | Amazon Technologies, Inc. | Personalized recommendations based on item usage |
| CN103345695A (en) * | 2013-06-25 | 2013-10-09 | 百度在线网络技术（北京）有限公司 | Commodity recommendation method and device |
| CN103577473A (en) * | 2012-08-03 | 2014-02-12 | 纽海信息技术（上海）有限公司 | Classification and disambiguation method, classification and disambiguation device and system thereof |
| CN103679494A (en) * | 2012-09-17 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Commodity information recommendation method and device |
| 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 |
| CN105809479A (en) * | 2016-03-07 | 2016-07-27 | 海信集团有限公司 | Item recommending method and device |
-
2016
- 2016-08-23 CN CN201610712391.9A patent/CN106372961A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8452797B1 (en) * | 2011-03-09 | 2013-05-28 | Amazon Technologies, Inc. | Personalized recommendations based on item usage |
| CN103577473A (en) * | 2012-08-03 | 2014-02-12 | 纽海信息技术（上海）有限公司 | Classification and disambiguation method, classification and disambiguation device and system thereof |
| CN103679494A (en) * | 2012-09-17 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Commodity information recommendation method and device |
| CN103345695A (en) * | 2013-06-25 | 2013-10-09 | 百度在线网络技术（北京）有限公司 | Commodity recommendation method and device |
| 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 |
| CN105809479A (en) * | 2016-03-07 | 2016-07-27 | 海信集团有限公司 | Item recommending method and device |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108628863A (en) * | 2017-03-15 | 2018-10-09 | 百度在线网络技术（北京）有限公司 | Information acquisition method and device |
| CN107341272A (en) * | 2017-08-25 | 2017-11-10 | 北京奇艺世纪科技有限公司 | A kind of method for pushing, device and electronic equipment |
| CN110069540A (en) * | 2017-09-08 | 2019-07-30 | 北京京东尚科信息技术有限公司 | Information output method and device |
| CN108090807A (en) * | 2017-12-13 | 2018-05-29 | 北京小度信息科技有限公司 | Information recommendation method and device |
| US10922737B2 (en) | 2017-12-22 | 2021-02-16 | Industrial Technology Research Institute | Interactive product recommendation method and non-transitory computer-readable medium |
| CN108154420A (en) * | 2017-12-26 | 2018-06-12 | 泰康保险集团股份有限公司 | Products Show method and device, storage medium, electronic equipment |
| CN110134469A (en) * | 2018-02-09 | 2019-08-16 | 优视科技有限公司 | Switching method, device and the computer equipment of holiday theme |
| CN108304115A (en) * | 2018-02-12 | 2018-07-20 | 广东欧珀移动通信有限公司 | Terminal entry selection method, device and storage medium and terminal |
| CN108053305B (en) * | 2018-02-12 | 2021-07-30 | 北京星选科技有限公司 | Data processing method and device applied to merchant promotion |
| CN108304115B (en) * | 2018-02-12 | 2020-12-04 | Oppo广东移动通信有限公司 | Terminal item selection method and device, storage medium and terminal |
| CN108053305A (en) * | 2018-02-12 | 2018-05-18 | 北京小度信息科技有限公司 | The data processing method and device promoted applied to trade company |
| CN108198051A (en) * | 2018-03-01 | 2018-06-22 | 口碑(上海)信息技术有限公司 | Across the Method of Commodity Recommendation and device of merchandise classification |
| CN108959340A (en) * | 2018-03-30 | 2018-12-07 | 深圳春沐源控股有限公司 | Page display method and page display system |
| CN110519318A (en) * | 2018-05-22 | 2019-11-29 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN110519318B (en) * | 2018-05-22 | 2023-08-04 | 北京京东尚科信息技术有限公司 | Information pushing method and device |
| CN108932530A (en) * | 2018-06-29 | 2018-12-04 | 新华三大数据技术有限公司 | The construction method and device of label system |
| CN109242649A (en) * | 2018-10-31 | 2019-01-18 | 广州品唯软件有限公司 | A kind of Method of Commodity Recommendation and relevant apparatus |
| CN109635153B (en) * | 2018-11-22 | 2023-01-13 | 腾讯音乐娱乐科技（深圳）有限公司 | Migration path generation method, device and storage medium |
| CN109635153A (en) * | 2018-11-22 | 2019-04-16 | 腾讯音乐娱乐科技（深圳）有限公司 | Migration path generation method, device and storage medium |
| CN111695922A (en) * | 2019-03-15 | 2020-09-22 | 北京沃东天骏信息技术有限公司 | Potential user determination method and device, storage medium and electronic equipment |
| CN110111186A (en) * | 2019-05-10 | 2019-08-09 | 拉扎斯网络科技（上海）有限公司 | Information processing method, device, electronic equipment and computer readable storage medium |
| CN110838043A (en) * | 2019-11-05 | 2020-02-25 | 智者四海（北京）技术有限公司 | Commodity recommendation method and device |
| CN111009299A (en) * | 2019-12-17 | 2020-04-14 | 武汉海云健康科技股份有限公司 | Similar medicine recommendation method and system, server and medium |
| CN111199459A (en) * | 2019-12-30 | 2020-05-26 | 深圳市盟天科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
| CN111861623A (en) * | 2019-12-30 | 2020-10-30 | 北京骑胜科技有限公司 | Information recommendation method, device and equipment |
| CN111199459B (en) * | 2019-12-30 | 2024-02-09 | 深圳市盟天科技有限公司 | Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium |
| CN113379516A (en) * | 2021-08-12 | 2021-09-10 | 永正信息技术(南京)有限公司 | Recommended product determination method and device |
| CN113610608A (en) * | 2021-08-19 | 2021-11-05 | 创优数字科技(广东)有限公司 | User preference recommendation method and device, electronic equipment and storage medium |
| CN115471302A (en) * | 2022-11-14 | 2022-12-13 | 山东智豆数字科技有限公司 | Electronic marketing data processing method based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106372961A (en) | Commodity recommendation method and device | |
| CN109509054B (en) | Commodity recommendation method under mass data, electronic device and storage medium | |
| EP3049912B1 (en) | Method and system for extracting user behavior features to personalize recommendations | |
| US20180165745A1 (en) | Intelligent Recommendation Method and System | |
| US20150379609A1 (en) | Generating recommendations for unfamiliar users by utilizing social side information | |
| US20160171539A1 (en) | Inference-Based Behavioral Personalization and Targeting | |
| US9082086B2 (en) | Adaptively learning a similarity model | |
| CN107169052A (en) | Recommend method and device | |
| CN105469263A (en) | Commodity recommendation method and device | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN105335409A (en) | Target user determination method and device and network server | |
| CN109242654A (en) | A kind of item recommendation method and system | |
| CN105096152A (en) | Commodity popularity-based operation execution method and device | |
| CN112818250B (en) | Content recommendation method, device, electronic equipment and system | |
| WO2016141197A1 (en) | Model-based replacement part system | |
| US20120296776A1 (en) | Adaptive interactive search | |
| CN106296270A (en) | Method of Commodity Recommendation and device | |
| CN107644036A (en) | A kind of method, apparatus and system of data object push | |
| CN106484766A (en) | Searching method based on artificial intelligence and device | |
| US20210256548A1 (en) | Computation of user-specific item-related values on an electronic processing platform | |
| CN114880566A (en) | User behavior analysis method, device, equipment and medium based on graph neural network | |
| CN111340522A (en) | Resource recommendation method, device, server and storage medium | |
| CN112749323A (en) | Method and device for constructing user portrait | |
| CN111489196B (en) | Prediction method and device based on deep learning network, electronic equipment and medium | |
| US10366088B2 (en) | Method and system for mining frequent and in-frequent items from a large transaction database |
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: 20170201 |