Commodity recommendation method based on user portrait

Abstract

The invention discloses a commodity recommendation method based on a user portrait. The method comprises the following steps: generating purchase preferences and purchase capabilities of a user based on purchase records and browsing records of the user; generating a commodity demand list of the user based on the purchase record and the browsing record; generating a commodity recommendation list of the user based on the commodity demand list and the purchase preference; generating a social relationship of the user based on the historical payment information of the user; based on the purchase preference, the purchase capability is determined. Generating an online shopping recommendation model according to the commodity demand list, the commodity recommendation list and the social relation; and recommending commodities to the user based on the online shopping recommendation model. According to the method, commodity recommendation is further carried out based on the online shopping time period of the user and the location of the user.

Description

Technical Field

[0001] The present application relates to the fields of artificial intelligence (AI) and big data, and more specifically, to a method for recommending products based on user portraits, and more specifically, to a method for accurately recommending products based on user portraits of user attributes such as user online shopping habits and social models. Background Art

[0002] User portrait is a labeled user model abstracted from the user's social attributes, living habits, consumption behavior habits and other user attribute information. Its purpose is to establish descriptive label attributes for the user in multiple dimensions. In the prior art, it is possible to collect data on multiple dimensions such as user's social attributes, consumption habits, preference characteristics, etc. through big data technology, and then characterize the user or product feature attributes, analyze and count these features, and mine potential value information, thereby abstracting the user's full information picture.

[0003] Big data technology has reached a relatively mature stage. Moreover, its application has become an indispensable part of the development of the Internet, and user portrait is one of the important applications of big data technology. With the continuous development of information technology, user portrait has been widely used in many aspects of Internet process optimization, targeted message and advertising push, user personalized service and improvement, etc., and can provide users with more efficient and targeted information delivery and user experience that is closer to personal habits.

[0004] In the current e-commerce industry, the method of constructing user portraits is generally to save the logs of users' browsing behaviors such as visiting product categories on the site, and then traverse all user behavior logs within a certain time window, and perform comprehensive calculations on them according to a specific weight decay function to obtain the latest user portrait. However, in the existing method of constructing user portrait models, each user is treated individually and independently, and the user's social relationship network is not comprehensively considered. Therefore, such user portraits are bound to be one-sided, which easily leads to unsatisfactory results when using user portraits for Internet applications. For example, if based on this construction method, product recommendations made to users often have the problem of low accuracy, resulting in a very low subsequent purchase success rate.

[0005] Therefore, there is a great need in the art for accurate product recommendation methods based on precise user portraits. Summary of the invention

[0006] As mentioned above, since the user portrait construction method in the prior art does not take the user's social attributes into consideration, it is impossible to construct an accurate user portrait, which ultimately leads to the inability to make accurate product recommendations based on the user portrait.

[0007] In order to solve the above problems, the present invention aims to provide a commodity recommendation method based on user portraits, which can accurately push commodities based on user portraits of many user attributes such as but not limited to user online shopping, social models, etc. This method can fully consider many user attributes that are not used in traditional recommendation models, such as the user's social relationship network, the specific time period of the user's online shopping, and the city where the user is located, and provide users with more accurate commodity push information, thereby greatly improving the user experience.

[0008] A brief summary of one or more aspects is given below to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all conceived aspects, and is neither intended to identify the key or critical elements of all aspects nor to define the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a prelude to a more detailed description given later.

[0009] The present application provides a commodity recommendation method based on user portrait, the method comprising: generating the user's purchase preference and purchasing ability based on the user's purchase record and browsing record; generating the user's commodity demand list based on the purchase record and the browsing record; generating the user's commodity recommendation list based on the commodity demand list and the purchase preference; generating the user's social relationship based on the user's historical payment information; generating an online shopping recommendation model based on the purchase preference, the purchasing ability, the commodity demand list, the commodity recommendation list and the social relationship; and recommending commodities to the user based on the online shopping recommendation model.

[0010] According to a preferred embodiment of the present application, generating the user's purchase preferences and purchasing capabilities based on the user's purchase records and browsing records further includes: selecting the top M preference tags in the purchase records and browsing records as the user's purchase preferences; and selecting the top N value tags in the purchase records and browsing records as the user's purchasing capabilities.

[0011] According to a preferred embodiment of the present application, generating the user's product recommendation list based on the product demand list and the purchase preference further includes: finding other similar products according to the category labels of the products listed in the product demand list, and filtering similar products according to the user's purchase preference to generate the product recommendation list.

[0012] According to a preferred embodiment of the present application, generating the user's social relationships based on the user's historical payment information further includes: analyzing the intimacy and social relationships between the user and other users based on the user's historical payment information and the user's account number to generate the user's social relationship network.

[0013] According to a preferred embodiment of the present application, the method further includes: based on the social relationship of the user, adding commodities listed in the demand lists of other users who have close relationships with the user to the user recommendation list.

[0014] According to a preferred embodiment of the present application, the method further includes: obtaining the online shopping period of the user based on the purchase record and browsing record of the user; and recommending products to the user based on the online shopping recommendation model and the online shopping period.

[0015] According to a preferred embodiment of the present application, the method further includes: obtaining the location of the user based on the IP of the user; and recommending products to the user based on the online shopping recommendation model and hot products in the location of the user.

[0016] According to a preferred embodiment of the present application, M is 5 and N is 3.

[0017] According to a preferred embodiment of the present application, the method further includes: recommending a delivery point closer to the city based on the user's location to reduce logistics time.

[0018] According to a preferred embodiment of the present application, the method further includes: establishing a social relationship model based on the user's social relationship network; recommending people who may know each other according to the social relationship model; and recommending similar products purchased by friends according to the social relationship model.

[0019] To achieve the foregoing and related ends, the one or more aspects include the features fully described below and particularly pointed out in the appended claims. The following description and drawings set forth in detail certain illustrative features of the one or more aspects. However, these features are merely indicative of several of the various ways in which the principles of the various aspects may be employed, and this description is intended to cover all such aspects and their equivalents. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to understand in detail the manner in which the above-stated features of the present application are used, a more specific description of the above briefly summarized contents may be made with reference to various aspects, some of which are illustrated in the accompanying drawings. However, it should be noted that the accompanying drawings illustrate only certain typical aspects of the present application and should not be considered to limit its scope, as the present description may allow for other equally effective aspects.

[0021] In the attached picture:

[0022] Figure 1 is a flowchart illustrating a method for recommending products based on user portraits according to an embodiment of the present application; and

[0023] Figure 2 It is a flowchart illustrating a method for recommending products based on user portraits according to a preferred embodiment of the present application. DETAILED DESCRIPTION

[0024] The detailed description set forth below in conjunction with the accompanying drawings is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. This detailed description includes specific details to provide a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well-known components are shown in block diagram form in order to avoid diluting such concepts.

[0025] It is to be understood that other embodiments would be apparent based on the present disclosure, and that system, structural, process, or mechanical changes may be made without departing from the scope of the present disclosure.

[0026] Specifically, compared with the prior art, the product recommendation method based on user portrait according to the present application mainly takes into account the following important aspects.

[0027] First, the number of preference tags and value tags corresponding to the user's online shopping records and browsing records is counted, and the top M preference tags are selected as the user's purchase preference and the top N value tags are selected as the user's purchasing power. For example, M can be 5 and N can be 3.

[0028] Of course, as those skilled in the art appreciate, the specific values ​​of M and N can be randomly selected by those skilled in the art based on specific needs, and these values ​​fall within the protection scope of this application.

[0029] Second, based on the user's purchase preferences, the user's online shopping browsing history is filtered to find products that match the purchase preferences and the user's online shopping history as the user's product demand list.

[0030] Third, based on the products listed in the product demand list, other similar products are found according to the category labels of the products, and similar products are filtered according to the user's purchasing preferences to obtain a product recommendation list.

[0031] Fourth, based on the user's historical payment information, such as historical payment information and the user's online shopping account information, the social relationship between the user and other users is analyzed to generate the user's social relationship network.

[0032] Fifth, generate a user online shopping model based on the user's purchasing preferences, economic capabilities, demand lists, recommendation lists, social relationship networks, etc.

[0033] Sixth, analyze the user's online shopping time period based on the user's shopping behavior, and push product information that matches the user's online shopping model to the user based on the time period.

[0034] Seventh, the city where the user is located is analyzed based on the IP address used by the user to access the Internet, so as to accurately push hot products in that city that match the user's online shopping model. This allows accurate push of products and their rapid delivery to users, greatly improving user experience and purchase success rate.

[0035] As the only accompanying drawing of this application, Figure 1 A flowchart of a method for recommending products based on user portraits according to an embodiment of the present invention is explained.

[0036] Reference Figure 1 , aspects are described with reference to one or more methods that can perform the actions described herein. Figure 1 The operations described in the present invention are presented in a particular order and / or as being performed by example components, but it should be understood that the order of these actions and the components performing the actions may vary depending on the implementation. In addition, it should be understood that the following actions or functions may be performed by a specially programmed processor, a processor executing specially programmed software or computer-readable media, or by any other combination of hardware components and / or software components capable of performing the described actions or functions.

[0037] like Figure 1 As shown in , according to the method 100 for accurately pushing products based on the user's historical online shopping behavior and social model of the present application, it comprehensively considers many parameters such as the user's historical online shopping consumption behavior and the user's social relationship attributes to provide the user with a more accurate online shopping recommendation service, thereby greatly improving the user experience.

[0038] First, in step S101, the user's purchase preference and purchasing power parameters are generated according to the user's purchase record and browsing record.

[0039] For example, if the user has frequently purchased baby products, milk powder, diapers, etc. online recently, it is inferred that the user has a lot of needs for newborn products recently and prefers to buy related products. For example, the top M preference tags are selected as the user's purchase preferences: milk powder, diapers, baby shampoo, etc. Of course, as those skilled in the art can appreciate, the number of M can be selected according to the actual needs of the model and is not limited to the data listed above.

[0040] Moreover, according to the level of the items purchased by the user among similar items, the purchasing power parameters of the user can be basically judged, such as high-end consumers, end consumers, or lower-end consumers. For example, the top N value tags with the highest value can be selected as the user's purchasing power. For example, if 0-200 yuan is set as tag 1; 200-500 is set as tag 2; and 500-1000 is set as tag 3. If the user purchases five items, 3 of which are valued as tag 1 and 2 are valued as tag 2, then the value tag with the largest number is tag 1. Similarly, the other N-1 value tags can be similarly counted. Therefore, the top N value tags with the highest value can be used as the user's purchasing power.

[0041] Of course, similarly, the value of N can be randomly selected by those skilled in the art based on specific needs, and these values ​​fall within the protection scope of this application.

[0042] Next, in step S102, commodities are screened out from the user's browsing history and purchase history to form a commodity demand list.

[0043] For example, filter out the following from the user's browsing and purchasing records in the past week: Brand A platinum milk powder stage 1, Brand B diapers size S, and Brand C baby shampoo.

[0044] Now go to step S103, find similar products according to the category labels of the above products, and filter similar products according to the user's purchase preferences to obtain a recommended product list.

[0045] Next, in step S104, the intimacy and social relationship between the user and other users are analyzed based on the user's purchase record, payment information (such as historical payment information), user account, etc., so as to generate a social relationship network of the specific user.

[0046] Then, in step S105, an online shopping recommendation model is generated based on the user's purchase preference, purchasing power, demand list, recommendation list, and social relationship.

[0047] For example, products in the demand list of users with whom the user has a preset relationship type may be added to the user's recommendation list, and the preset relationship includes, but is not limited to, friend relationship, lover relationship, family relationship, etc.

[0048] For example, if the preset relationship type is a family relationship, then in this embodiment, the recommendation of products in each other's demand list is mainly recommended to the family. For example, if the user's family wants to buy a handbag recently, the demand will be added to the user's recommendation list, and the user will be regularly reminded and recommended.

[0049] Finally, in step S106, suitable products are recommended to the user based on the generated online shopping recommendation model.

[0050] It should be noted that in this embodiment, the commodities added to the user's recommendation list correspond to the user's purchasing power, so as to avoid adding commodities beyond the user's economic ability, thereby making the recommended commodity list more accurate.

[0051] In addition, the method according to the present application can also establish a social relationship model for the user based on the user's social relationship network, and then recommend people they may know based on the social relationship model; and recommend similar products that their friends have purchased based on the social relationship model.

[0052] Figure 2 A further preferred embodiment of the present application is shown in FIG. 1 , in which a product recommendation method 100 ′ based on user portrait is shown.

[0053] like Figure 2 As shown in , after the online shopping recommendation model is generated at step S105, the success rate of recommendation and the transaction rate of goods can be improved according to the generated user online shopping model, user online shopping time period, user's city, etc.

[0054] For example, based on the generation of the online shopping recommendation model, as shown in step S107, the user's online shopping time period is analyzed according to the user's shopping behavior, and product information matching the user's online shopping model can be pushed to the user during a determined time period every day (step S109).

[0055] Furthermore, according to the method of the present application, in a more preferred embodiment, as shown in step S108, the city where the user is located is analyzed based on the IP address of the user's Internet access, so as to accurately push the hot commodities in the city that match the user's online shopping model, thereby further realizing the accurate push of commodities (step S109).

[0056] In this way, not only the success rate of users purchasing goods is improved, but also users have a very good shopping experience.

[0057] In summary, in the technical solution of the present application, the social relationship between users is analyzed through the user's online shopping behavior, account data and other data, so that the user's social relationship network and other user attributes are included in the final generated user online shopping model, which can make the final user portrait model more accurate. At the same time, by using intimacy and social relationships, the products in the demand list of other users who are close to the user, such as the user's relatives, spouse, friends, etc., are pushed to the user, so that the user can understand the recent needs of his relatives and friends. On the one hand, when the user chooses gifts for relatives, spouses or friends, the user can choose goods that are "to their liking", which not only reduces the difficulty of the user to choose gifts, but also ensures that the selected gifts are really needed and liked by the other party, thereby enhancing their intimacy. In addition, people who may know the user can be recommended to increase their social scope, and goods bought by relatives and friends can also be recommended to them.

[0058] In addition, by analyzing the user's online shopping time period and the user's Internet IP address, we can understand the user's online shopping habits and the recent hot commodities in the city where he is located, so that we can push notifications to the user in the corresponding time period about the hot commodities in the city and the list that meets the user's needs. Therefore, by combining user portraits, accurate recommendations can be achieved, which is not only conducive to improving the success rate of recommendations and the transaction rate of commodities, but also greatly improves the user's shopping experience.

[0059] According to various aspects, elements, or any part of elements, or any combination of elements of the present disclosure, a "processing system" including one or more processors can be implemented. Examples of processors include: microprocessors, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other suitable hardware configured to perform various functionalities described throughout the present disclosure. One or more processors in a processing system can execute software. Software should be broadly interpreted as meaning instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc., whether it is described in software, firmware, middleware, microcode, hardware description language, or other terms. Software may reside on a computer-readable medium. The computer-readable medium may be a non-transient computer-readable medium. As examples, non-transitory computer-readable media include: magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips), optical disks (e.g., compact disks (CDs), digital versatile disks (DVDs)), smart cards, flash memory devices (e.g., memory cards, memory sticks, key drives), random access memory (RAM), read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), registers, removable disks, and any other suitable medium for storing software and / or instructions that can be accessed and read by a computer. As examples, computer-readable media may also include carrier waves, transmission lines, and any other suitable medium for transmitting software and / or instructions that can be accessed and read by a computer. The computer-readable medium may reside in the processing system, be external to the processing system, or be distributed across multiple entities including the processing system. The computer-readable medium may be implemented in a computer program product. As examples, a computer program product may include a computer-readable medium in packaging materials. Those skilled in the art will recognize how to best implement the described functionality presented throughout this disclosure depending on the specific application and the overall design constraints imposed on the overall system.

[0060] It should be understood that the specific order or hierarchy of steps in the disclosed methods is an illustration of an exemplary process. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the methods or methodologies described herein may be rearranged. The attached method claims present elements of the various steps in a sample order and are not meant to be limited to the specific order or hierarchy presented unless specifically stated herein.

[0061] The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the universal principles defined herein may be applied to other aspects. Therefore, the claims are not intended to be limited to the various aspects shown herein, but should be granted the full scope consistent with the language of the claims, wherein the singular reference to the element is not intended to mean "there is and only one" (unless specifically stated) but "one or more". Unless specifically stated otherwise, the term "some" refers to one or more. The phrase "at least one" quoting a list of items refers to any combination of these items, including a single member. As an example, "at least one of a, b or c" is intended to cover: at least one a; at least one b; at least one c; at least one a and at least one b; at least one a and at least one c; at least one b and at least one c; and at least one a, at least one b and at least one c. The elements of the various aspects described throughout this disclosure are all structurally and functionally equivalent schemes currently or hereafter known to those of ordinary skill in the art, and are expressly incorporated herein by reference, and are intended to be covered by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims.

Claims

1. A commodity recommendation method based on user portrait, characterized in that: The method comprises: Generating the user's purchase preference and purchasing ability based on the user's purchase record and browsing record; Generating a commodity demand list of the user based on the purchase record and the browsing record; Generating a commodity recommendation list for the user based on the commodity demand list and the purchase preference; generating a social relationship of the user based on the historical payment information of the user; generating an online shopping recommendation model based on the purchase preference, the purchase ability, the commodity demand list, the commodity recommendation list and the social relationship; and Recommending products to the user based on the online shopping recommendation model.

2. The commodity recommendation method according to claim 1, characterized in that: Generating the user's purchase preference and purchasing ability based on the user's purchase record and browsing record further includes: Selecting the top M preference tags in the purchase records and the browsing records as the purchase preferences of the user; and The top N value tags in the purchase records and the browsing records are selected as the purchasing ability of the user.

3. The commodity recommendation method according to claim 1, characterized in that: Generating the user's commodity recommendation list based on the commodity demand list and the purchase preference further includes: Other commodities of the same type are found according to the category labels of the commodities listed in the commodity demand list, and the commodities of the same type are screened according to the purchase preference of the user to generate the commodity recommendation list.

4. The commodity recommendation method according to claim 1, characterized in that: Generating the user's social relationships based on the user's historical payment information further includes: The intimacy and social relationship between the user and other users are analyzed based on the user's historical payment information and the user's account number to generate the user's social relationship network.

5. The commodity recommendation method according to claim 1 or 4, characterized in that: The method further comprises: Based on the social relationship of the user, commodities listed in the demand lists of other users who have close relationships with the user are added to the user recommendation list.

6. The commodity recommendation method according to claim 1 or 4, characterized in that: The method further comprises: Obtaining the online shopping period of the user based on the purchase record and browsing record of the user; Recommending products to the user based on the online shopping recommendation model and the online shopping period.

7. The commodity recommendation method according to claim 6, characterized in that: The method further comprises: Obtaining the location of the user based on the user's IP; and Recommending products to the user based on the online shopping recommendation model and hot products in the user's location.

8. The commodity recommendation method according to claim 2, characterized in that: M is 5 and N is 3.

9. The commodity recommendation method according to claim 7, characterized in that: The method further comprises: Based on the user's location, a shipping point closer to the city is recommended to reduce logistics time.

10. The commodity recommendation method according to claim 4, characterized in that: The method further comprises: Establishing a social relationship model based on the social relationship network of the user; Recommending people you may know based on the social relationship model; and Recommend similar products purchased by friends based on the social relationship model.

Images