CN111144986A - Commodity recommendation method and device for social e-commerce website based on sharing behavior - Google Patents
Commodity recommendation method and device for social e-commerce website based on sharing behavior Download PDFInfo
- Publication number
- CN111144986A CN111144986A CN201911360358.4A CN201911360358A CN111144986A CN 111144986 A CN111144986 A CN 111144986A CN 201911360358 A CN201911360358 A CN 201911360358A CN 111144986 A CN111144986 A CN 111144986A
- Authority
- CN
- China
- Prior art keywords
- commodity
- model
- user
- information
- social
- 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
Images
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
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The embodiment of the invention provides a social e-commerce website commodity recommendation method and device based on sharing behaviors, wherein the method comprises the following steps: acquiring commodity sharing information; inputting the commodity sharing information into a preset commodity recommendation model to obtain a commodity recommendation result; the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label. Establishing a fusion model through a sharer influence model and the sharee influence model, obtaining a user interest vector, a user social influence vector and a commodity characterization vector in the fusion model through learning according to a fusion model loss function, deducing the probability that all commodities are purchased by friends after being recommended by a user, and sequencing in sequence to obtain a commodity recommendation result.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to a social e-commerce website commodity recommendation method and device based on sharing behaviors.
Background
With the development of internet technology, Electronic Business (E-Business) is seen everywhere in our daily life and work. Electronic commerce generally refers to a novel business operation mode for implementing online shopping of consumers, online transactions between merchants and online electronic payments, as well as various business activities, transaction activities, financial activities and related comprehensive service activities based on the internet in a network environment where the internet is open, among the wide variety of business activities worldwide.
The social e-commerce is a subset of e-commerce and comprises social media and online media functions, social behaviors among users can be supported, so that online shopping and commodity selling services are assisted, and different from the traditional e-commerce in which the users shop through searching and intelligent recommendation, the users of the social e-commerce can share links of commodities to friends of the users, and can also directly click the links to purchase. Different from the traditional recommendation system, the social recommendation system aims to consider the social relationship among users so as to assist in modeling the user interest and improve the recommendation effect.
However, the existing social e-commerce website recommendation method does not consider the social influence, and the recommendation is not accurate enough, so how to more effectively recommend the social e-commerce website commodity recommendation based on the sharing behavior has become a problem to be solved in the industry.
Disclosure of Invention
The embodiment of the invention provides a social e-commerce website commodity recommendation method and device based on sharing behaviors, which are used for solving the technical problems in the background technology or at least partially solving the technical problems in the background technology.
In a first aspect, an embodiment of the present invention provides a social e-commerce website commodity recommendation method based on a sharing behavior, including:
acquiring commodity sharing information;
inputting the commodity sharing information into a preset commodity recommendation model to obtain a commodity recommendation result;
the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
More specifically, before the step of inputting the commodity sharing information and the sharing social relationship into a preset commodity recommendation model, the method further includes:
acquiring user commodity sample data with an interactive tag;
randomly acquiring user commodity sample data with a label which is not interacted from an unobserved sample set, and constructing positive and negative sample pair information according to the user commodity sample data with the label which is interacted and the user commodity sample data with the label which is not interacted;
and training the preset commodity recommendation model by using the positive and negative samples, and finishing training when the preset training conditions are met to obtain the preset commodity recommendation model.
More specifically, before the step of inputting the positive and negative sample pair information into a commodity recommendation model for training, the method further includes:
constructing a sharer influence model according to the user influence representation information and the commodity representation information;
constructing an influence model of the sharee according to the representation information of the sharee and the commodity representation information;
obtaining a fusion model according to the sharer influence model and the sharee influence model;
and optimizing the fusion model through a fusion model loss function to obtain a preset commodity recommendation model.
More specifically, the step of optimizing the fusion model through the fusion model loss function to obtain the preset commodity recommendation model specifically includes:
acquiring user commodity sample information with shared purchase labels and user commodity sample information with free purchase labels to construct positive and negative sample pair information;
training a fusion model loss function parameter according to the positive and negative sample pair information through a gradient random reduction method, and when a preset training condition is met, stabilizing the fusion model loss function, so that a social-common electronic commerce platform fusion model is obtained according to the fusion model loss function.
In a second aspect, an embodiment of the present invention provides a social e-commerce website commodity recommendation device based on a sharing behavior, including:
the acquisition module is used for acquiring user sharing information;
the recommendation module is used for inputting the commodity sharing information into a preset commodity recommendation model to obtain a commodity recommendation result;
the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
In a third aspect, an embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method for recommending social e-commerce website commodities based on sharing behaviors as described in the first aspect when executing the program.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method for recommending social e-commerce website commodities based on sharing behaviors as described in the first aspect.
According to the social e-commerce website commodity recommendation method and device based on the sharing behaviors, a fusion model is established through a sharer influence model and a sharee influence model, a user interest vector, a user social influence vector and a commodity characterization vector in the fusion model are obtained through learning according to a fusion model loss function, the probability that all commodities are purchased by friends after the commodities are recommended by users is deduced, and the commodities are sequenced in sequence to obtain a commodity recommendation result.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart illustrating a method for recommending commodities to a social network site based on sharing behavior according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a social network site commodity recommendation device based on sharing behavior according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart of a method for recommending commodities to a social network site based on sharing behavior according to an embodiment of the present invention, as shown in fig. 1, including:
step S1, commodity sharing information is obtained;
step S2, inputting the commodity sharing information and the sharing social relationship into a preset commodity recommendation model to obtain a commodity recommendation result;
the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
Specifically, the social e-commerce platform described in the embodiment of the present invention refers to an e-commerce platform in which a social relationship exists and mutual influence exists between users, the existence of the social relationship is mainly different between the social e-commerce platform and a general e-commerce platform, the user commodity interaction behavior data described in the embodiment of the present invention refers to interaction records of the users on the social e-commerce platform, and the social relationship data refers to friend relationships of the users on the social e-commerce platform.
The user commodity interaction behavior data described in the embodiment of the invention specifically refers to interaction behavior data of a user and a merchant, and comprises behavior data of clicking commodities by the user and behavior data of collecting or purchasing the commodities by the user.
The commodity sharing information described in the embodiment of the invention refers to information that a user shares commodities to friends.
After a user selects a friend to be shared, the preset commodity recommendation model deduces the probability of purchasing all commodities by the friends according to commodity sharing information and user interest vectors, user social influence vectors and commodity characterization vectors obtained through model training and learning, and carries out sequencing in sequence, and a platform selects a plurality of commodities most likely to be purchased by the friends according to sequencing results to carry out recommendation display.
The shared purchase tag described in the embodiment of the present invention refers to a tag of a commodity that is purchased by a user after receiving recommendation information of a friend.
The free purchase tag described in the embodiment of the invention is a commodity tag which is not recommended by friends, is not influenced by social contact and is purchased by the user according to the subjective intention.
The preset commodity recommendation model described in the embodiment of the invention is obtained by training according to the commodity sample information of the user with the shared purchase label and the commodity sample information of the user with the free purchase label.
According to the embodiment of the invention, a fusion model is established through a sharer influence model and the sharee influence model, a user interest vector, a user social influence vector and a commodity representation vector in the fusion model are obtained through learning according to a fusion model loss function, the probability that all commodities are purchased by friends after being recommended by a user is inferred, and the commodity recommendation results are obtained through sequencing in sequence.
On the basis of the above embodiment, before the step of inputting the commodity sharing information and the sharing social relationship into a preset commodity recommendation model, the method further includes:
acquiring user commodity sample data with an interactive tag;
randomly acquiring user commodity sample data with a label which is not interacted from an unobserved sample set, and constructing positive and negative sample pair information according to the user commodity sample data with the label which is interacted and the user commodity sample data with the label which is not interacted;
and training the preset commodity recommendation model by using the positive and negative samples, and finishing training when the preset training conditions are met to obtain the preset commodity recommendation model.
Specifically, the interactive tag described in the embodiment of the present invention represents that the user product sample data is data interacted with the user, and the non-interactive tag refers to that the user product sample data is data that has not been interacted with the user.
The user commodity sample data with the interactive tag is obtained from a user log collected by a social e-commerce platform, a personalized commodity list is generated for a user, and only data sets of the user and the commodity existing in the two platforms are reserved, so that the user commodity sample data with the interactive tag is obtained.
The unobserved sample set described in the embodiment of the present invention specifically refers to data that has not interacted with a user, and the constructing of the positive and negative sample pair information specifically refers to that based on the positive: and the proportion of minus 1:1 takes the user commodity sample data with the interactive label as a positive example, and takes the user commodity sample data without the interactive label as a negative example, so as to construct the user commodity sample data with the interactive label and the user commodity sample data without the interactive label.
The preset training condition described in the embodiment of the present invention may refer to a preset number of training rounds, for example, 300 training rounds, or may refer to a preset training time, for example, a training time of 30 minutes.
According to the embodiment of the invention, the association between the social influence and the interest of the user during sharing is established, so that the cooperative learning can be effectively realized from the two parts of behavior data in a joint learning manner, so that a model can be better fitted, and better model parameters can be obtained.
On the basis of the above embodiment, before the step of inputting the positive and negative sample pair information into a commodity recommendation model for training, the method further includes:
constructing a sharer influence model according to the user influence representation information and the commodity representation information;
constructing an influence model of the sharee according to the representation information of the sharee and the commodity representation information;
obtaining a fusion model according to the sharer influence model and the sharee influence model;
and optimizing the fusion model through a fusion model loss function to obtain a preset commodity recommendation model.
Specifically, the sharer influence model described in the embodiment of the present invention specifically means that, in the social e-commerce, the user behavior is likely to be influenced by the sharer. For example, a friend of the user may know about clothing and then share certain clothing with the user, and the user may likely purchase the clothing. That is, for each sharing behavior of the user, the influence of the sharer on the sharee needs to be characterized.
Constructing a sharer influence model according to the user influence representation information and the commodity representation information, and specifically comprising the following steps:
wherein s isuToken vector for user u as sharer, qiThe commodity characterization information is obtained.
The greater the influence here means that the item i shared by the user u to his friends is more likely to be purchased by his friends.
For each sharing action, the interest of the user receiving the sharing is another important factor for deciding whether to take the purchasing action. If the user is not interested in the merchandise, the probability that the user purchases the merchandise will not be high even if the user is shared with a high social influence. According to the sharee representation information and the commodity representation information, an influence model of the sharee is constructed, and the method specifically comprises the following steps:
wherein q isiCharacterizing information for a commodity, qiThe commodity characterization information is obtained.
The more influential the sharee is, the more probability that the user v receiving the share of the commodity i generates the purchasing behavior, and the probability is only related to the interest of the user v.
The fusion model specifically comprises:
wherein s isuAnd puRespectively as token vectors of sharer and sharee for user u, qiFor the merchandise characterization information, α is a controlled interest and social influence association mechanism hyper-parameter, pvThe vector is characterized for user interest.
On the basis of the above embodiment, the step of optimizing the fusion model through the fusion model loss function to obtain the preset commodity recommendation model specifically includes:
acquiring user commodity sample information with shared purchase labels and user commodity sample information with free purchase labels to construct positive and negative sample pair information;
training a fusion model loss function parameter according to the positive and negative sample pair information through a gradient random reduction method, and when a preset training condition is met, stabilizing the fusion model loss function, so that a social-common electronic commerce platform fusion model is obtained according to the fusion model loss function.
The invention provides a common optimization sharing and subsequent purchasing behavior log and a free purchasing log of a user, and particularly, optimization targets of the two parts are combined to be used as a unified optimization target. Specifically, the following loss function is adopted:
here, share and share represent share and subsequent purchase behavior and free purchase behavior, respectively.
Wherein
And
the triple data which are constructed according to the method and are used for partial order learning and share the follow-up purchasing behavior and the free interaction behavior respectively,
and
respectively is the estimated probability difference value of an interactive sample and a non-interactive sample sharing the following purchase behavior and the partial order behavior of the free interactive behavior, j represents a negative sample obtained by sampling, Reg is a regular term for solving overfitting and is PThe L-2 norm of two matrices Q (P is one column of P and Q is one column of Q) is multiplied by the sum of a constant (called the regularization term coefficient, which generally ranges from 1e-6 to 1e-2, depending on the particular experiment).
The above formula also gives the amount of parameter update per model iteration: Δ pu,Δpv,ΔqiAnd Δ qjAfter a certain number of model iterations, the loss function L will stabilize at a smaller value, and the model effect cannot be improved by continuing the training at this time, so the model training is ended, the parameters are not updated any more, and the model at this time is output as the final model.
On the basis of the above embodiment, the user logs of the social commerce platform from 20 days in 2017, 9 months to 22 days in 2017, 10 months and 22 are utilized to construct the commodity recommendation system for the sharing scene. Data set correlation statistics are shown in table 1:
TABLE 1
| Number of users who initiated the sharing | 175,827 |
| Number of users who have accepted the sharing | 380,639 |
| Number of shared commodities | 75,464 |
| Number of shared purchases | 574,273 |
| Number of users who have made over-free purchases | 3,011,253 |
| Producing the number of commodities purchased over time | 1,231,307 |
| Number of records of free purchase | 43,273,133 |
| Number of users who have generated two interactions | 40,011 |
| Number of commodities for which two interactions have occurred | 46,340 |
In the above statistics, the number of users determines the size of the model parameters: the relevant model parameters of the users exist in a matrix form, the dimensionality of the matrix is the multiplication of the number of the users by the dimensionality of the hidden space, and the commodity is similar to the hidden space dimensionality. And the number of the divided behavior records determines the number of the training and testing samples. Specifically, to ensure that both interactions exist, the end-use numbers of users and merchandise are 40,011 and 46,340.
Firstly, training data is constructed, for two kinds of interactive data, users and commodities with less than five records are removed to ensure reliable data set division, then the records are sorted according to the time stamps, the record with the latest time is used as a test set, and the rest records are used as training sets.
An Adam optimizer is selected as an optimizer for random gradient descent, and due to the over-fitting problem possibly existing in the dot product model, regular terms are introduced to all the characterization matrixes to prevent over-fitting. The coefficients are all searched within [1e-1,1e-2,1e-3,1e-4,1e-5,1e-6,1e-7,1e-8] for the regularization term, within [8,16,32,64,128,256,512] for the hidden space dimension closely related to the model capacity, and a more appropriate 0.0001 is selected for the learning rate hyperparameter of the model Adam optimizer. And the total iteration times are set to 5000 times so as to ensure the convergence of the model, and the optimal hyper-parameter combination is judged and selected as the final model according to the sequencing performance on the test set.
On the social e-commerce platform, an entrance for a user to make commodity recommendation to friends of the user is provided. After the user selects the friends expected to be shared, the model output by the method deduces the probability of purchasing all commodities by the friends through the user interest vector, the user social influence vector and the commodity characterization vector obtained by learning, and carries out sequencing in sequence, and the platform selects a plurality of commodities most likely to be purchased by the friends according to the sequencing result to carry out recommendation and display.
Fig. 2 is a schematic structural diagram of a sharing-behavior-based social e-commerce website commodity recommendation device according to an embodiment of the present invention, as shown in fig. 2, including: an acquisition module 210 and a recommendation module 220; the obtaining module 210 is configured to obtain user sharing information and sharing social information; the recommendation module 220 is configured to input the commodity sharing information and the sharing social relationship into a preset commodity recommendation model to obtain a commodity recommendation result; the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
The apparatus provided in the embodiment of the present invention is used for executing the above method embodiments, and for details of the process and the details, reference is made to the above embodiments, which are not described herein again.
According to the embodiment of the invention, a fusion model is established through a sharer influence model and the sharee influence model, a user interest vector, a user social influence vector and a commodity representation vector in the fusion model are obtained through learning according to a fusion model loss function, the probability that all commodities are purchased by friends after being recommended by a user is inferred, and the commodity recommendation results are obtained through sequencing in sequence.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 3, the electronic device may include: a processor (processor)310, a communication Interface (communication Interface)320, a memory (memory)330 and a communication bus 340, wherein the processor 310, the communication Interface 320 and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call logic instructions in the memory 330 to perform the following method: acquiring commodity sharing information; inputting the commodity sharing information and the sharing social relationship into a preset commodity recommendation model to obtain a commodity recommendation result; the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
In addition, the logic instructions in the memory 330 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
An embodiment of the present invention discloses a computer program product, which includes a computer program stored on a non-transitory computer readable storage medium, the computer program including program instructions, when the program instructions are executed by a computer, the computer can execute the methods provided by the above method embodiments, for example, the method includes: acquiring commodity sharing information; inputting the commodity sharing information and the sharing social relationship into a preset commodity recommendation model to obtain a commodity recommendation result; the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
Embodiments of the present invention provide a non-transitory computer-readable storage medium storing server instructions, where the server instructions cause a computer to execute the method provided in the foregoing embodiments, for example, the method includes: acquiring commodity sharing information; inputting the commodity sharing information and the sharing social relationship into a preset commodity recommendation model to obtain a commodity recommendation result; the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A social e-commerce website commodity recommendation method based on sharing behaviors is characterized by comprising the following steps:
acquiring commodity sharing information;
inputting the commodity sharing information into a preset commodity recommendation model to obtain a commodity recommendation result;
the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
2. The social e-commerce website commodity recommendation method based on sharing behavior according to claim 1, wherein before the step of inputting the commodity sharing information into a preset commodity recommendation model, the method further comprises:
constructing positive and negative sample pair information according to the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label;
and training the preset commodity recommendation model by using the positive and negative samples, and finishing training when the preset training conditions are met to obtain the preset commodity recommendation model.
3. The social e-commerce website commodity recommendation method based on sharing behavior of claim 2, wherein before the step of training the positive and negative sample pair information input into a commodity recommendation model, the method further comprises:
constructing a sharer influence model according to the user influence representation information and the commodity representation information;
constructing an influence model of the sharee according to the representation information of the sharee and the commodity representation information;
obtaining a fusion model according to the sharer influence model and the sharee influence model;
and optimizing the fusion model through a fusion model loss function to obtain a preset commodity recommendation model.
4. The social e-commerce website commodity recommendation method based on the sharing behavior as claimed in claim 3, wherein the fusion model is specifically as follows:
wherein s isuAnd puRespectively as token vectors of sharer and sharee for user u, qiFor the merchandise characterization information, α is a controlled interest and social influence association mechanism hyper-parameter, pvThe vector is characterized for user interest.
5. The social e-commerce website commodity recommendation method based on the sharing behavior as claimed in claim 4, wherein the step of optimizing the fusion model through a fusion model loss function to obtain a preset commodity recommendation model specifically comprises:
acquiring user commodity sample information with shared purchase labels and user commodity sample information with free purchase labels to construct positive and negative sample pair information;
training a fusion model loss function parameter according to the positive and negative sample pair information through a gradient random reduction method, and when a preset training condition is met, stabilizing the fusion model loss function, so that a social-common electronic commerce platform fusion model is obtained according to the fusion model loss function.
6. The utility model provides a social merchant website commodity recommendation device based on share action which characterized in that includes:
the acquisition module is used for acquiring commodity sharing information;
the recommendation module is used for inputting the commodity sharing information into a preset commodity recommendation model to obtain a commodity recommendation result;
the preset commodity recommendation model is based on the user commodity sample information with the shared purchase label and the user commodity sample information with the free purchase label.
7. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for recommending social electronic commerce website commodities based on sharing behaviors as claimed in any one of claims 1 to 5 when executing the program.
8. A non-transitory computer readable storage medium, having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method for social merchant website merchandise recommendation based on sharing behavior according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911360358.4A CN111144986A (en) | 2019-12-25 | 2019-12-25 | Commodity recommendation method and device for social e-commerce website based on sharing behavior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911360358.4A CN111144986A (en) | 2019-12-25 | 2019-12-25 | Commodity recommendation method and device for social e-commerce website based on sharing behavior |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111144986A true CN111144986A (en) | 2020-05-12 |
Family
ID=70520143
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911360358.4A Pending CN111144986A (en) | 2019-12-25 | 2019-12-25 | Commodity recommendation method and device for social e-commerce website based on sharing behavior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111144986A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112328779A (en) * | 2020-11-04 | 2021-02-05 | 中国平安人寿保险股份有限公司 | Training sample construction method and device, terminal equipment and storage medium |
| CN112437349A (en) * | 2020-11-10 | 2021-03-02 | 杭州时趣信息技术有限公司 | Video stream recommendation method and related device |
| CN113379494A (en) * | 2021-06-10 | 2021-09-10 | 清华大学 | Commodity recommendation method and device based on heterogeneous social relationship and electronic equipment |
| CN113868542A (en) * | 2021-11-25 | 2021-12-31 | 平安科技（深圳）有限公司 | Attention model-based push data acquisition method, device, equipment and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130246186A1 (en) * | 2012-03-15 | 2013-09-19 | Buide Limited | Social shopping platform providing recommender display and advertisement publishing and shopping method thereof |
| WO2017157149A1 (en) * | 2016-03-15 | 2017-09-21 | 平安科技（深圳）有限公司 | Social network-based recommendation method and apparatus, server and storage medium |
| CN107369058A (en) * | 2016-05-13 | 2017-11-21 | 华为技术有限公司 | A kind of correlation recommendation method and server |
| CN110008399A (en) * | 2019-01-30 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of training method and device, a kind of recommended method and device of recommended models |
| CN110222272A (en) * | 2019-04-18 | 2019-09-10 | 广东工业大学 | A kind of potential customers excavate and recommended method |
-
2019
- 2019-12-25 CN CN201911360358.4A patent/CN111144986A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130246186A1 (en) * | 2012-03-15 | 2013-09-19 | Buide Limited | Social shopping platform providing recommender display and advertisement publishing and shopping method thereof |
| WO2017157149A1 (en) * | 2016-03-15 | 2017-09-21 | 平安科技（深圳）有限公司 | Social network-based recommendation method and apparatus, server and storage medium |
| CN107369058A (en) * | 2016-05-13 | 2017-11-21 | 华为技术有限公司 | A kind of correlation recommendation method and server |
| CN110008399A (en) * | 2019-01-30 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of training method and device, a kind of recommended method and device of recommended models |
| CN110222272A (en) * | 2019-04-18 | 2019-09-10 | 广东工业大学 | A kind of potential customers excavate and recommended method |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112328779A (en) * | 2020-11-04 | 2021-02-05 | 中国平安人寿保险股份有限公司 | Training sample construction method and device, terminal equipment and storage medium |
| CN112328779B (en) * | 2020-11-04 | 2024-02-13 | 中国平安人寿保险股份有限公司 | Training sample construction method, device, terminal equipment and storage medium |
| CN112437349A (en) * | 2020-11-10 | 2021-03-02 | 杭州时趣信息技术有限公司 | Video stream recommendation method and related device |
| CN113379494A (en) * | 2021-06-10 | 2021-09-10 | 清华大学 | Commodity recommendation method and device based on heterogeneous social relationship and electronic equipment |
| CN113868542A (en) * | 2021-11-25 | 2021-12-31 | 平安科技（深圳）有限公司 | Attention model-based push data acquisition method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111144986A (en) | Commodity recommendation method and device for social e-commerce website based on sharing behavior | |
| CN111008858B (en) | Commodity sales prediction method and system | |
| CN109615454A (en) | Determine the method and device of user's finance default risk | |
| CN112435067A (en) | Intelligent advertisement putting method and system for cross-e-commerce platform and social platform | |
| CN110197390B (en) | Recommendation method and system based on association degree and economic value of association rule | |
| CN107301592A (en) | The method and device excavated for commodity substitute | |
| US20130204701A1 (en) | Apparatus, system and methods for marketing targeted products to users of social media | |
| CN111695965A (en) | Product screening method, system and equipment based on graph neural network | |
| WO2020221022A1 (en) | Service object recommendation method | |
| CN111177581A (en) | Multi-platform-based social e-commerce website commodity recommendation method and device | |
| US20160005072A1 (en) | Method and system for online commerce analysis | |
| US20210109906A1 (en) | Clustering model analysis for big data environments | |
| KR102307662B1 (en) | Product information provision system and method thereof | |
| Koppenberg et al. | Markup estimation: A comparison of contemporary methods at the example of European food retailers | |
| CN112801803B (en) | Financial product recommendation method and device | |
| CN113610521A (en) | Method and apparatus for detecting anomalies in behavioral data | |
| CN112330373A (en) | User behavior analysis method and device and computer readable storage medium | |
| CN111275480B (en) | Multi-dimensional sparse sales data warehouse oriented fraud behavior mining method | |
| CN117094764A (en) | Bank integral processing method and device | |
| CN111915344A (en) | New member ripening accelerating method and device based on medical big data | |
| CN111639274B (en) | Online commodity intelligent sorting method, device, computer equipment and storage medium | |
| Ram et al. | Digital technology adoption behaviour in the context of unorganised retail: towards a technology continuance theory | |
| CN113516496B (en) | Advertisement conversion rate estimation model construction method, device, equipment and medium thereof | |
| CN111639989B (en) | Commodity recommendation method and readable storage medium | |
| Pandey et al. | Recommending e-commerce products on cold start and long tail using transaction data |
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 |