Product recommendation method and device based on user portrait, medium and electronic equipment

Abstract

The invention discloses a product recommendation method and device based on a user portrait, a medium and electronic equipment, and relates to the technical field of privacy computing. The method comprises the steps that a user portrait of a target user is acquired, the user portrait at least comprises one of a user interest portrait, a user group portrait and a user attribute portrait, the user interest portrait is the user portrait constructed based on user behavior data, and the user attribute portrait is the user portrait constructed based on user attribute data; the user group portrait is a user portrait constructed based on user behavior data of a plurality of users; and based on the user portrait, determining a recommended product of the target user. The recommendation system based on the user portrait is constructed through multi-dimensional user analysis, the personalized recommended product list is generated for the user, the recommendation accuracy is improved, meanwhile, user experience is considered, excessive recommendation or improper recommendation of products is avoided, and the user satisfaction degree is improved.

Description

Technical Field

[0001] The present invention relates to the field of privacy computing technology, and in particular to a product recommendation method, device, medium and electronic device based on user portrait. Background Art

[0002] The basic idea of ​​the content-based recommendation algorithm is to select other objects with similar features from the recommended objects as recommendation results based on the objects that the user has already selected. This recommendation strategy first extracts the content features of the recommended objects and matches them with the user's interest preferences in the user model. The recommended objects with a higher matching degree can be recommended to the user as recommendation results, but this is prone to "over-recommendation" phenomenon.

[0003] The recommendation algorithm based on collaborative filtering is a classic recommendation algorithm. When the system has user preference information, user usage or purchase history, and user historical rating data, it can mine users' potential and unknown interests and hobbies, and show good recommendation performance. However, with the complexity of content, the increase of users and projects, the algorithm itself has some problems, such as cold start problem and data sparsity problem.

[0004] There is currently no effective solution to the problems of "over-recommendation", cold start and data sparsity in recommendation algorithms in related technologies. Summary of the invention

[0005] The main purpose of this application is to provide a product recommendation method, device, medium and electronic device based on user portrait to solve the problems of "over-recommendation", cold start and data sparsity existing in the recommendation algorithm in the related art.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a product recommendation method based on user portrait is provided. The method comprises: obtaining a user portrait of a target user, wherein the user portrait comprises at least one of the following: a user interest portrait, a user group portrait, and a user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users; based on the user portrait, determining a recommended product for the target user.

[0007] Optionally, the user portrait includes a user interest portrait, and the recommended products for the target user are determined based on the user portrait, including the following steps: based on the user portrait, multiple interest tags and multiple tag weights of the target user are determined, the tag weights are used to evaluate the importance of the interest tags, and the interest tags and tag weights are set in a one-to-one correspondence; based on the interest tags and tag weights, the target user's interest recommended products are determined, and the interest recommended products are products under the interest tag category.

[0008] Optionally, the user portrait includes a user group portrait, and determining the recommended products for the target user based on the user portrait includes the following steps: determining the target user group based on the user group portrait, the target user group being the user group to which the target user belongs; obtaining group purchase product data of the target user group; and determining group recommended products for the target user based on the group purchase product data.

[0009] Optionally, the user portrait includes a user attribute portrait, and determining the recommended products for the target user based on the user portrait includes the following steps: determining a reference user based on the user attribute portrait, wherein the user attribute portrait of the reference user is set identically to the user attribute portrait of the target user; obtaining user behavior data of the reference user; and determining a reference recommended product for the target user based on the user behavior data of the reference user.

[0010] Optionally, before obtaining the user portrait of the target user, the method also includes: obtaining a user interest model of the target user and user behavior data of the target user, the user behavior data including at least one of the following: user click data, user browsing data, user search history data, user purchase record data, user evaluation data, and user feedback data; based on the user behavior data, determining user interest characteristics of the target user, the user interest characteristics including at least one of the following: visited product categories, product purchase frequency, and product ratings; using the user interest model, analyzing the user interest characteristics, and generating a user interest portrait of the target user.

[0011] Optionally, before obtaining the user portrait of the target user, the method also includes: obtaining user behavior data of multiple users; dividing the multiple users into at least one user group based on the multiple user behavior data; extracting group interest characteristics of the user group, the group interest characteristics including at least one of the following: interest product category; generating a user group portrait based on the group interest characteristics.

[0012] Optionally, before obtaining the user portrait of the target user, the method also includes: obtaining user attribute data of the target user, the user attribute data including at least one of the following: age, gender, occupation, educational background, geographic location, and income; based on the user attribute data, extracting attribute characteristics of the target user; based on the attribute characteristics, generating a user attribute portrait of the target user.

[0013] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a product recommendation device based on user portrait is provided. The device comprises: an acquisition module, the acquisition module is used to acquire the user portrait of the target user, the user portrait comprises at least one of the following: user interest portrait, user group portrait, user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users; a determination module, the determination module is used to determine the recommended product for the target user based on the user portrait.

[0014] According to another aspect of the present application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned product recommendation method based on user portrait.

[0015] According to another aspect of the present application, an electronic device is provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the above-mentioned product recommendation method based on user portrait is executed when the program is running.

[0016] According to another aspect of the present application, a computer program product is also provided, including computer instructions, which implement the steps of the above-mentioned product recommendation method based on user portrait when executed by a processor.

[0017] In the embodiment of the present application, a user portrait of the target user is obtained, and the user portrait includes at least one of the following: user interest portrait, user group portrait, and user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users; based on the user portrait, the recommended products for the target user are determined. By analyzing users in multiple dimensions, a recommendation system based on user portraits is constructed to generate a personalized list of recommended products for the target user, thereby improving the accuracy of the recommendation, while taking into account the user experience, avoiding over-recommendations or recommending inappropriate products, maintaining user trust, improving user satisfaction, promoting product sales, and improving the overall business of the financial industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 A hardware structure block diagram of a computer terminal for implementing a product recommendation method based on user portrait is shown;

[0020] Figure 2 is a flowchart of a product recommendation method based on user portraits provided according to an embodiment of the present application;

[0021] Figure 3 is a schematic diagram of the structure of a product recommendation device based on user portraits provided according to an embodiment of the present application;

[0022] Figure 4 is a flowchart of a product recommendation method based on user interest portraits provided according to an embodiment of the present application;

[0023] Figure 5 is a flowchart of a hierarchical clustering algorithm provided according to an embodiment of the present application;

[0024] Figure 6 is a flowchart of a product recommendation method based on user group portraits provided according to an embodiment of the present application;

[0025] Figure 7 is a flowchart of a product recommendation method based on user attribute portraits provided according to an embodiment of the present application;

[0026] Figure 8 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following explanations:

[0030] User portrait: User portrait refers to a comprehensive description and characterization of users based on their various attributes, behaviors, preferences and other information, in order to better understand users, predict user behavior, and provide users with personalized services and experiences.

[0031] Hierarchical clustering algorithm: an iterative algorithm for unsupervised learning that aims to organize data points hierarchically into a tree structure to reflect the similarities between data.

[0032] Collaborative filtering algorithm: A well-known and commonly used recommendation algorithm that discovers user preferences based on mining user historical behavior data, and predicts products that users may like and makes recommendations.

[0033] TF-IDF (term frequency-inverse document frequency) algorithm: A statistical method used to evaluate the importance of a word to a document set or a document in a corpus.

[0034] Over-recommendation: Giving excessive advice or recommendations in a situation, even to the point of annoying others.

[0035] Cold start problem: In a recommendation system, the ability to recommend new users or new items is limited or insufficient.

[0036] Data sparsity problem: There is a large amount of missing data in the dataset, which means that in a given dataset, only a few entries or features have valid values, while other entries or features may be missing values ​​or zero values. This situation may affect tasks such as data analysis, model building, and recommendation systems, because the lack of sufficient data will reduce the performance and accuracy of the model.

[0037] It should be noted that the acquisition of user portraits involved in this application, and the user data based on the user portraits (including but not limited to data for display, data for analysis, etc.) are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with relevant laws, regulations and standards, and necessary confidentiality measures are taken, which do not violate public order and good customs, and corresponding operation entrances are provided for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions, providing users with corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making; if the user chooses to refuse, the expert decision-making process will be entered.

[0038] Example 1

[0039] According to an embodiment of the present application, an embodiment of a product recommendation method based on user portrait is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0040] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal (or mobile device) for implementing a product recommendation method based on user portrait is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0041] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0042] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the product recommendation method based on user portrait in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizing the above-mentioned product recommendation method based on user portrait. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0043] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0044] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0045] Under the above operating environment, this application provides Figure 2 The product recommendation method based on user portrait is shown. Figure 2 1 is a flowchart of a product recommendation method based on user portrait according to Example 1 of the present application. The product recommendation method based on user portrait comprises the following steps:

[0046] Step S21, obtaining a user portrait of the target user, the user portrait includes at least one of the following: a user interest portrait, a user group portrait, and a user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users;

[0047] Specifically, the user interest portrait is a user portrait constructed based on the different interests of the target users. The user interest portrait is reflected through the user's behavioral data, such as the target user's browsing history, click records, purchase records, favorites, likes, tag classification and other data on different financial products; the user group portrait is a user portrait that represents the characteristics of a group or a category of users by analyzing and clustering the common hobbies and preference feature data of the group. The user group portrait can mine the product types that are favored by similar users based on the clustering algorithm; the user attribute portrait is a user portrait model constructed by collecting and analyzing the personal attribute data of the target users. The user attribute portrait can be constructed based on data such as the user's gender, age, income, region, behavior, habits and psychological characteristics.

[0048] It should be noted that, when obtaining the user portrait of the target user, you can obtain only one user portrait, or you can obtain multiple user portraits at the same time. For example, you can finally establish and construct a final user portrait based on the user interest portrait, user group portrait, and user attribute portrait, and comprehensively determine the user portrait of the target user through multiple user portraits.

[0049] Step S22, determining recommended products for target users based on the user portrait.

[0050] Specifically, establish a user portrait, determine the recommendation system based on the user portrait, and recommend suitable personalized financial products to target users.

[0051] Through the above steps, the user portrait of the target user is obtained. The user portrait includes at least one of the following: user interest portrait, user group portrait, and user attribute portrait, among which the user interest portrait is a user portrait built based on user behavior data, the user attribute portrait is a user portrait built based on user attribute data, and the user group portrait is a user portrait built based on the user behavior data of multiple users; based on the user portrait, the recommended products for the target user are determined. Through multi-dimensional analysis of users, a recommendation system based on user portraits is constructed to generate a personalized list of recommended products for the target user, thereby improving the accuracy of the recommendation. At the same time, the user experience is considered to avoid over-recommendation or recommendation of inappropriate products, maintain user trust, improve user satisfaction, promote product sales, and enhance the overall business of the financial industry.

[0052] Optionally, in step S21, the user portrait includes a user interest portrait, and based on the user portrait, determining the recommended product for the target user includes the following steps:

[0053] Step S211, based on the user portrait, determining multiple interest tags and multiple tag weights of the target user, the tag weights are used to evaluate the importance of the interest tags, and the interest tags and tag weights are set in a one-to-one correspondence;

[0054] Specifically, the user interest portrait is constructed by labeling and classifying multiple interests and hobbies of the target user, and weighting each interest tag, wherein the interest tag can be ranged according to all browsing history data of the target user, and the tag weight can be determined according to the number of times and browsing time of different products by the target user.

[0055] It should be noted that the label weight of a financial product can be determined by the number of times the target user browses or clicks on a financial product and the time the product is browsed. In this way, according to the target user's interest range and the number of times and time the products of interest are browsed, the products are ranked according to their importance in the target user's mind.

[0056] Those skilled in the art should understand that the target user's label weight for a financial product is mainly determined by the number of times the product is viewed and the browsing time, but the label weight can also be revised based on factors such as the depth and breadth of the user's understanding of the product and time decay, and the label weight is ultimately normalized to ensure that the sum of the weights of all labels is equal to 1, which is convenient for subsequent calculations.

[0057] Step S212, based on the interest tags and tag weights, determine the target user's interest recommended products, where the interest recommended products are products classified under the interest tags.

[0058] Specifically, different products are sorted according to interest tags, and products classified under the interest tags are recommended to target users in sequence or all at once.

[0059] In this embodiment, by setting interest tags and tag weights, user interests can be understood more carefully, and products that best match the interest preferences of target users can be recommended, thereby improving the personalization of recommendations, user satisfaction and business conversion rate. In the financial industry, it can be applied to personalized recommendations for financial products, insurance products, and banking services, helping users quickly find investment and financial services that are suitable for them.

[0060] Optionally, in step S21, the user portrait includes a user group portrait, and based on the user portrait, determining the recommended product for the target user includes the following steps:

[0061] Step S213, determining a target user group based on the user group portrait, where the target user group is the user group to which the target user belongs;

[0062] Specifically, data is collected and analyzed on the purchase records of multiple target users, the products purchased by the target users are classified, and then the target users are classified into groups to determine the target user groups. The target users are classified based on user group portraits, and the target users are classified into the same or similar product target user groups based on the target users' historical data of purchasing and browsing products. That is, the target user groups are classified according to the same interests according to the target users' historical data.

[0063] Step S214, obtaining group purchase product data of the target user group;

[0064] Specifically, the data on product purchases by each target user in the target user group is comprehensively collected, including data such as the types of products purchased by different target users, the quantity of products, the types of products browsed and the browsing time.

[0065] Step S215, determining group recommended products for target users based on group product purchase data.

[0066] Specifically, the products purchased by each target user are not exactly the same, so the target user group recommended products can be determined based on the data of the products purchased by the target user group, and the group recommended products are all the product categories purchased by the target user group. Among them, the product categories are sorted according to the number of products purchased by the target user group in the group recommended products, and the products can be recommended to the target user in order.

[0067] It should be noted that, since the types of products purchased by each target user are different, other products that have not been purchased by the target user who has not purchased all the products in the target user group can be recommended to the target user who has not purchased all the products in the target user group; products can also be recommended to the novice target user as a whole. That is, the potential preference types of financial products for individual target users in the target user group are mined, and the group wisdom is used to provide users with products that may be of interest but have not yet been exposed to.

[0068] Optionally, in step S21, the user portrait includes a user attribute portrait, and based on the user portrait, determining the recommended product for the target user includes the following steps:

[0069] Step S216, determining a reference user based on the user attribute profile, wherein the user attribute profile of the reference user is set to be identical to the user attribute profile of the target user;

[0070] Specifically, an attribute profile of the target user is determined, and based on the attribute profile, a reference user with the same attribute profile as the target user is determined according to the same or similar data matching. The attribute data may be a combination of one or more data including gender, age, income, region, and habits.

[0071] It should be noted that the more types of data that are combined based on which the attribute portrait is determined, the closer the reference user is to the target user, the higher the reference value provided by the reference user, and the more the products recommended to the target user for purchase by the reference user will meet the needs of the target user.

[0072] Step S217, obtaining user behavior data of the reference user;

[0073] Specifically, after determining the reference user, the behavioral data of the reference user is obtained and analyzed. The behavioral data of the reference user includes the reference user's transaction record, browsing and search history, product click-through rate and dwell time, risk assessment record and interactive feedback evaluation data.

[0074] Step S218, determining a reference recommended product for the target user based on the user behavior data of the reference user.

[0075] Specifically, based on the products purchased, browsed, and understood by the reference users, these products are listed as reference recommended products for the target users.

[0076] In this embodiment, the financial product recommendation based on the user attribute profile describes the overall preferences of users in the same income range, age and region, providing users with a good reference for financial products. In addition, the financial product recommendation method in this embodiment does not have the cold start problem compared to the product recommendation method in the prior art. Even if the target user has never generated any purchase record, the product can be recommended based on the target user's attribute profile.

[0077] Optionally, in step S21, before obtaining the user portrait of the target user, the method further includes:

[0078] Step S200, obtaining a user interest model of a target user and user behavior data of the target user, wherein the user behavior data includes at least one of the following: user click data, user browsing data, user search history data, user purchase record data, user evaluation data, and user feedback data;

[0079] Specifically, after obtaining the user behavior data of the target user, the user behavior data is integrated and a user interest model is constructed based on the user behavior data. The user interest model is a data model that depicts the user's interest portrait and the user's psychological portrait.

[0080] Among them, based on user behavior data, cluster analysis, collaborative filtering, deep learning and other algorithms are used to capture the dynamic changes of user interests and build a user interest model. The user interest model can realize the function of personalized product recommendations for target users, and can also help understand the target users' current needs for different financial products, services and information.

[0081] Step S201, determining the user interest characteristics of the target user based on the user behavior data, where the user interest characteristics include at least one of the following: visited product category, product purchase frequency, and product rating;

[0082] Specifically, the user behavior data of the target users are integrated and analyzed, and the interest characteristics of the target users are extracted. The target users' interest focuses and interest changes can be extracted through user behavior pattern analysis and user interest time series analysis. In addition, the users' complex interest patterns can be mined through collaborative filtering, text mining, and deep learning models.

[0083] Step S202: Analyze user interest characteristics using the user interest model and generate a user interest portrait of the target user.

[0084] Specifically, a user interest model is generated based on user behavior data, user interest features are extracted through the user interest model, and the user's basic attributes are combined with the interest features obtained through model analysis to generate a personalized user interest portrait of the target user.

[0085] It should be noted that the target user has multiple interest characteristics. When combining the interest characteristics of the target user with the basic attributes of the target user, the weighted average method can also be used to combine them to ensure that the weights of different interest characteristics can reflect their importance in the overall interest of the user. After the interest portrait of the target user is generated, the portrait can be further refined according to different business scenarios to facilitate more refined product recommendations for the target user.

[0086] Optionally, in step S21, before obtaining the user portrait of the target user, the method further includes:

[0087] Step S203, obtaining user behavior data of multiple users;

[0088] Specifically, the user behavior data generated by multiple target users are collected, analyzed, and stored, and the user behavior data of each target user includes at least one of the following data: user click data, user browsing data, user search history data, user purchase record data, user evaluation data, and user feedback data.

[0089] Step S204, dividing the multiple users into at least one user group based on the multiple user behavior data;

[0090] Specifically, the characteristics of each target user group are analyzed to identify the commonalities within the group. Based on the analysis results, multiple target users are divided into a user group according to a certain commonality within the group, and the characteristics of each target user group are described. Based on multiple user behavior data, multiple users can be divided into the same user group or multiple different user groups.

[0091] Step S205, extracting the group interest characteristics of the user group, the group interest characteristics including at least one of the following: interest product category;

[0092] Specifically, the basis for dividing user groups is the existence of a certain common interest feature. At least the commonality of the user groups can be extracted as the interest features of the user groups. The group interest features can also include the group's investment risk preference, financial management knowledge level, investment behavior patterns, investment goals, etc.

[0093] Step S206, generating a user group portrait based on the group interest characteristics.

[0094] Specifically, the group interest features of the user group are extracted. There may be multiple group interest features. The weight of the overall performance and importance in the group is determined for each group interest feature. The extracted group features are combined with the basic attributes of the group to generate a user group portrait.

[0095] In this embodiment, detailed user group portraits are generated based on group interest characteristics, which can better understand the target market, guide product innovation and formulate marketing strategies, thereby improving customer satisfaction and market competitiveness. During the use of user group portraits, tests are conducted to verify the construction strategies or marketing effects under different group portraits, and feature weights and portrait models are adjusted based on feedback to optimize group portraits to improve their accuracy and practicality.

[0096] Optionally, in step S21, before obtaining the user portrait of the target user, the method further includes:

[0097] Step S207, obtaining user attribute data of the target user, the user attribute data including at least one of the following: age, gender, occupation, educational background, geographic location, income;

[0098] Specifically, user attribute data is the personal information data of the target user, which is private and complex. User attribute data is closely related to the personality, psychology, and interest characteristics of the target user. The investment personality, investment psychology, and interest characteristics of the target user can be estimated through user attribute data.

[0099] Step S208, extracting attribute features of the target user based on the user attribute data;

[0100] Specifically, the user attribute data includes multiple user information, and the common features in the user information of the target user are filtered to extract the individual features of the target user, that is, feature selection can extract the corresponding attribute features of the target user. There can be multiple attribute features or one attribute feature.

[0101] Step S209: Generate a user attribute profile of the target user based on the attribute features.

[0102] Specifically, after extracting multiple attribute features of the target user, the distribution of each attribute feature is calculated, and the attribute features are analyzed using a machine learning algorithm to identify the most influential and accurate feature combination for predicting the target user's behavior, and establish a corresponding user attribute profile.

[0103] In this embodiment, the attribute characteristics of the user can be effectively extracted and quantified based on the user attribute data, and more accurate and personalized services can be provided to the user according to the attribute characteristics.

[0104] according to Figure 4 As shown, the present application also provides a preferred embodiment of a product recommendation method based on user interest portraits.

[0105] Specifically, the product recommendation method based on user interest portrait recommends financial products based on the interest tag weights obtained when the user browses the financial product data. The interest tag weights are obtained based on the user's purchase records and the relevant content of the financial product, and are related to the user's interests.

[0106] Specifically, the number of financial products of this type that need to be recommended can be obtained by multiplying the weight of the interest tag by the total number of financial products to be recommended (for example, the total number of financial products is 10). The algorithm formula is as follows:

[0107]

[0108] In formula (1), x j is the recommended quantity of products in category j; A j is the weight of the jth interest tag.

[0109] like Figure 4 As shown in the figure, based on the user portrait, a weight ratio model of interest tags is constructed, and the relevant number is recommended according to the weight of each tag to combine a recommended list of financial products.

[0110] Combination Figure 5 , Figure 6 As shown, the present application also provides a preferred embodiment of a product recommendation method based on user group portraits.

[0111] Specifically, users of the same type have similar interests. Recommending financial products to these users is likely to win their favor, or they will like the products after learning about them. This can help us mine users’ potential financial product preferences. The hierarchical clustering algorithm is used to generate interest profiles, such as Figure 5 As shown, the algorithm is implemented by the following steps:

[0112] Step 1: Get the purchase records of all users and save the detailed information of the financial products purchased by each user in the initial data text;

[0113] Step 2: Segment the initial data text according to the user and delete stop words. The obtained data is saved in the initial label text.

[0114] Step 3: Convert the original label text into a term matrix, and use the TF-IDF algorithm to calculate the weight of each word to obtain a weight matrix;

[0115] Step 4: Calculate the similarity between clusters by using the weighted matrix.

[0116] The similarity formula is as follows:

[0117]

[0118] In formula (2), the cosine similarity is the cosine of the angle between two n vectors in space, which is the dot product of the two vectors divided by the product of the lengths of the two vectors. i , B i They represent the components of vectors A and B respectively. When cosθ is closer to 1, the preferences of user groups A and B are more similar.

[0119]

[0120] In formula (3), A1 and A2 represent two clusters, C i and C j For the clusters in A1 and A2 respectively, an appropriate distance threshold is selected according to the results of system clustering, and clusters with similar distances are continuously merged according to the threshold until the distance between clusters reaches the set threshold, such as Figure 6 As shown, the clustering result is used as the input of the collaborative filtering model to query users in the same user group as the current user, and recommend to him the top ten financial products that other users have purchased the most but the user has not purchased.

[0121] Combination Figure 7 As shown, the present application also provides a preferred embodiment of a product recommendation method based on user attribute portrait.

[0122] Specifically, based on the fact that users with different incomes, ages, and regions have different preferences for financial products, the product recommendation method based on user attribute portraits selects income range, age, and customer business handling area as recommendation conditions, and uses the above attributes to classify users. At the same time, the idea of ​​collaborative filtering is also referred to for recommendation. The financial products that users with the same basic attribute portrait have seen the most are recommended, which can realize financial product recommendations based on user attribute portraits. The financial product recommendation based on user attribute portraits describes the overall preferences of users with the same income range, age, and region, and provides users with a good reference for financial products. In addition, compared with the financial product recommendation method used previously, the movie recommendation method based on user attribute portraits does not have a cold start problem. Even if the user has never generated any purchase records, financial products can be recommended based on the user's attribute portrait.

[0123] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0124] Example 2

[0125] The embodiment of the present application also provides a product recommendation device based on user portrait. It should be noted that the product recommendation device based on user portrait in the embodiment of the present application can be used to execute the product recommendation method based on user portrait provided in the embodiment of the present application. The product recommendation device based on user portrait provided in the embodiment of the present application is introduced below.

[0126] According to an embodiment of the present application, a device for implementing the above-mentioned product recommendation method based on user portrait is also provided, such as Figure 3 As shown, the device comprises:

[0127] An acquisition module 32, which is used to acquire a user portrait of a target user. The user portrait includes at least one of the following: a user interest portrait, a user group portrait, and a user attribute portrait. The user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users;

[0128] The determination module 34 is used to determine the recommended products for the target user based on the user portrait.

[0129] The product recommendation device based on user portrait provided in the embodiment of the present application obtains the user portrait of the target user, and the user portrait includes at least one of the following: user interest portrait, user group portrait, and user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users; based on the user portrait, the recommended products for the target user are determined. By analyzing users in multiple dimensions, a recommendation system based on user portraits is constructed to generate a personalized list of recommended products for the target user, thereby improving the accuracy of the recommendation, while taking into account the user experience, avoiding over-recommendations or recommendations of inappropriate products, maintaining user trust, improving user satisfaction, promoting product sales, and improving the overall business of the financial industry.

[0130] Optionally, in the product recommendation device based on user portrait provided in the present application, the determination module 34 includes a first determination module 341, and the first determination module 341 is used to: determine multiple interest tags and multiple tag weights of the target user based on the user portrait, the tag weights are used to evaluate the importance of the interest tags, and the interest tags and tag weights are set in a one-to-one correspondence; based on the interest tags and tag weights, determine the target user's interest recommended products, and the interest recommended products are products under the interest tag classification.

[0131] Optionally, in the product recommendation device based on user portrait provided in the present application, the determination module 34 includes a second determination module 342, and the second determination module 342 is used to: determine the target user group based on the user group portrait, the target user group is the user group to which the target user belongs; obtain the group purchase product data of the target user group; and determine the group recommended products for the target user based on the group purchase product data.

[0132] Optionally, in the product recommendation device based on user portrait provided in the present application, the determination module 34 includes a third determination module 343, and the third determination module 343 is used to: determine a reference user based on the user attribute portrait, and the user attribute portrait of the reference user is set the same as the user attribute portrait of the target user; obtain user behavior data of the reference user; and determine a reference recommended product for the target user based on the user behavior data of the reference user.

[0133] Optionally, in the product recommendation device based on user portrait provided in the present application, the product recommendation device also includes a first generation module 301, and the first generation module 301 is used to: obtain a user interest model of the target user and user behavior data of the target user, the user behavior data including at least one of the following: user click data, user browsing data, user search history data, user purchase record data, user evaluation data, user feedback data; based on the user behavior data, determine the user interest characteristics of the target user, the user interest characteristics including at least one of the following: visited product categories, product purchase frequency, product ratings; use the user interest model to analyze the user interest characteristics and generate a user interest portrait of the target user.

[0134] Optionally, in the product recommendation device based on user portrait provided in the present application, the product recommendation device also includes a second generation module 302, and the second generation module 302 is used to: obtain user behavior data of multiple users; based on the multiple user behavior data, divide the multiple users into at least one user group; extract group interest characteristics of the user group, and the group interest characteristics include at least one of the following: interest product category; based on the group interest characteristics, generate a user group portrait.

[0135] Optionally, in the product recommendation device based on user portrait provided in the present application, the product recommendation device also includes a third generation module 303, and the third generation module 303 is used to: obtain user attribute data of the target user, the user attribute data including at least one of the following: age, gender, occupation, educational background, geographic location, income; based on the user attribute data, extract attribute characteristics of the target user; based on the attribute characteristics, generate a user attribute portrait of the target user.

[0136] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0137] Optionally, the user portrait-based product recommendation device provided in the present application may also include other modules. For example, the user portrait-based product recommendation device may also include a communication module, which is used to transmit the signal data such as the user portrait acquired by the acquisition module 32 to the determination module 34 and the data storage device or the cloud in a wired or wireless manner; the user portrait-based product recommendation device may also include an optimization module, which continuously optimizes the user portrait-based product recommendation device and recommendation model by using a deep learning algorithm, and improves the accuracy and effect of the recommendation through a feedback loop.

[0138] It should be noted that the acquisition module 32 and the determination module 34 correspond to steps S21 to S22 in Example 1, and the examples and application scenarios implemented by the two modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n), and the above-mentioned modules may also be part of the device and may be run in the computer terminal 10 provided in Example 1.

[0139] Example 3

[0140] An embodiment of the present application may provide an electronic device, Figure 8 is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 8 As shown, the electronic device may include: one or more ( Figure 8 (only one is shown) processor 802, memory 804, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0141] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0142] The processor may call the information and application program stored in the memory through the transmission device to perform the following steps:

[0143] Step S21, obtaining a user portrait of the target user, the user portrait includes at least one of the following: a user interest portrait, a user group portrait, and a user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users;

[0144] Step S22, determining recommended products for target users based on the user portrait.

[0145] The processor may also call the information and application program stored in the memory through the transmission device to perform the following steps:

[0146] Step S211, based on the user portrait, determining multiple interest tags and multiple tag weights of the target user, the tag weights are used to evaluate the importance of the interest tags, and the interest tags and tag weights are set in a one-to-one correspondence;

[0147] Step S212, based on the interest tags and tag weights, determine the target user's interest recommended products, where the interest recommended products are products classified under the interest tags.

[0148] The processor may also call the information and application program stored in the memory through the transmission device to perform the following steps:

[0149] Step S213, determining a target user group based on the user group portrait, where the target user group is the user group to which the target user belongs;

[0150] Step S214, obtaining group purchase product data of the target user group;

[0151] Step S215, determining group recommended products for target users based on group product purchase data.

[0152] The processor may also call the information and application program stored in the memory through the transmission device to perform the following steps:

[0153] Step S216, determining a reference user based on the user attribute profile, wherein the user attribute profile of the reference user is set to be identical to the user attribute profile of the target user;

[0154] Step S217, obtaining user behavior data of the reference user;

[0155] Step S218, determining a reference recommended product for the target user based on the user behavior data of the reference user.

[0156] The processor may also call the information and application program stored in the memory through the transmission device to perform the following steps:

[0157] Step S200, obtaining a user interest model of a target user and user behavior data of the target user, wherein the user behavior data includes at least one of the following: user click data, user browsing data, user search history data, user purchase record data, user evaluation data, and user feedback data;

[0158] Step S201, determining the user interest characteristics of the target user based on the user behavior data, where the user interest characteristics include at least one of the following: visited product category, product purchase frequency, and product rating;

[0159] Step S202: Analyze user interest characteristics using the user interest model and generate a user interest portrait of the target user.

[0160] The processor may also call the information and application program stored in the memory through the transmission device to perform the following steps:

[0161] Step S203, obtaining user behavior data of multiple users;

[0162] Step S204, dividing the multiple users into at least one user group based on the multiple user behavior data;

[0163] Step S205, extracting the group interest characteristics of the user group, the group interest characteristics including at least one of the following: interest product category;

[0164] Step S206, generating a user group portrait based on the group interest characteristics.

[0165] The processor may also call the information and application program stored in the memory through the transmission device to perform the following steps:

[0166] Step S207, obtaining user attribute data of the target user, the user attribute data including at least one of the following: age, gender, occupation, educational background, geographic location, income;

[0167] Step S208, extracting attribute features of the target user based on the user attribute data;

[0168] Step S209: Generate a user attribute profile of the target user based on the attribute features.

[0169] By adopting the embodiment of the present application, a user portrait of the target user is obtained, and the user portrait includes at least one of the following: user interest portrait, user group portrait, and user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users; based on the user portrait, the recommended products for the target user are determined. By analyzing users in multiple dimensions, a recommendation system based on user portraits is constructed to generate a personalized list of recommended products for the target user, thereby improving the accuracy of the recommendation. At the same time, the user experience is considered to avoid over-recommendation or recommendation of inappropriate products, maintain user trust, improve user satisfaction, promote product sales, and enhance the overall business of the financial industry.

[0170] It can be understood by those skilled in the art that Figure 8The structure shown is for illustration only, and the electronic device may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, or other terminal devices. Figure 8 The structure of the electronic device is not limited. Figure 8 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 8 Different configurations are shown.

[0171] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0172] Example 4

[0173] The embodiment of the present application further provides a readable storage medium. Optionally, in this embodiment, the readable storage medium can be used to store the program code executed by the product recommendation method based on user portrait provided in the first embodiment.

[0174] Optionally, in this embodiment, the readable storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0175] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of a product recommendation method based on a user portrait.

[0176] Optionally, in this embodiment, the computer program is configured to perform the following steps:

[0177] Step S1, obtaining a user portrait of the target user, the user portrait includes at least one of the following: user interest portrait, user group portrait, user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users;

[0178] Step S2: Determine the recommended products for the target user based on the user portrait.

[0179] The readable storage medium provided in the present application stores a computer program, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0180] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0181] Step S1, obtaining a user portrait of the target user, the user portrait includes at least one of the following: user interest portrait, user group portrait, user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on user behavior data of multiple users;

[0182] Step S2: Determine the recommended products for the target user based on the user portrait.

[0183] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0184] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0185] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0186] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0187] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0188] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0189] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A product recommendation method based on user portrait, characterized in that: include: Obtain a user portrait of the target user, wherein the user portrait includes at least one of the following: a user interest portrait, a user group portrait, and a user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on the user behavior data of multiple users; Based on the user portrait, determine the recommended products for the target user.

2. The method according to claim 1, characterized in that The user portrait includes the user interest portrait. Based on the user portrait, determining the recommended product for the target user includes the following steps: Based on the user portrait, determining a plurality of interest tags and a plurality of tag weights of the target user, wherein the tag weights are used to evaluate the importance of the interest tags, and the interest tags and the tag weights are set in a one-to-one correspondence; Based on the interest tag and the tag weight, the interest recommended product of the target user is determined, and the interest recommended product is a product classified under the interest tag.

3. The method according to claim 1, characterized in that The user portrait includes the user group portrait. Based on the user portrait, determining the recommended product for the target user includes the following steps: Based on the user group portrait, determine a target user group, where the target user group is the user group to which the target user belongs; Acquire group purchase product data of the target user group; Based on the group product purchase data, group recommended products for the target users are determined.

4. The method according to claim 1, characterized in that: The user portrait includes the user attribute portrait. Based on the user portrait, determining the recommended product for the target user includes the following steps: Based on the user attribute profile, determining a reference user, wherein the user attribute profile of the reference user is set identically to the user attribute profile of the target user; Acquire the user behavior data of the reference user; Based on the user behavior data of the reference user, a reference recommended product for the target user is determined.

5. The method according to claim 2, characterized in that: Before obtaining the user portrait of the target user, the method further includes: Acquire the user interest model of the target user and the user behavior data of the target user, wherein the user behavior data includes at least one of the following: user click data, user browsing data, user search history data, user purchase record data, user evaluation data, and user feedback data; Determine the user interest characteristics of the target user based on the user behavior data, wherein the user interest characteristics include at least one of the following: visited product category, product purchase frequency, and product rating; The user interest model is used to analyze the user interest characteristics and generate a user interest portrait of the target user.

6. The method according to claim 5, characterized in that Before obtaining the user portrait of the target user, the method further includes: Obtaining the user behavior data of multiple users; Based on the plurality of user behavior data, the plurality of users are divided into at least one user group; Extracting group interest features of the user group, where the group interest features include at least one of the following: interest product categories; Based on the group interest characteristics, the user group portrait is generated.

7. The method according to claim 4, characterized in that Before obtaining the user portrait of the target user, the method further includes: Acquire user attribute data of the target user, wherein the user attribute data includes at least one of the following: age, gender, occupation, educational background, geographic location, and income; Based on the user attribute data, extracting attribute characteristics of the target user; Based on the attribute characteristics, a user attribute profile of the target user is generated.

8. A product recommendation device based on user portrait, characterized in that: The device comprises: An acquisition module, the acquisition module is used to acquire a user portrait of a target user, the user portrait includes at least one of the following: a user interest portrait, a user group portrait, and a user attribute portrait, wherein the user interest portrait is a user portrait constructed based on user behavior data, the user attribute portrait is a user portrait constructed based on user attribute data, and the user group portrait is a user portrait constructed based on the user behavior data of multiple users; A determination module is used to determine the recommended products for the target user based on the user portrait.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the product recommendation method based on user portrait according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program, when running, executes the product recommendation method based on user portrait described in any one of claims 1 to 7.

11. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by the processor, the steps of the product recommendation method based on user portrait described in any one of claims 1 to 7 are implemented.

Images