Product recommendation method and device, equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence and the field of financial services, in particular to a product recommendation method and device, equipment and a medium. The client basic information of the target client is acquired, the target client is portrayed, the initial client portrait is matched with the preset product library, the matched product is sent to the target client, the target client is portrayed again, and the product in the second matching result is sent to the target user as the recommended product. The method can be applied to an insurance product recommendation service scene, the target customer is determined to obtain the basic information to construct the initial customer portrait, the initial customer portrait is matched with the preset product library to obtain the first matching result, the authorization verification information is sent, the customer obtains the social information after the customer allows authorization, and the basic information is combined to obtain the optimized customer portrait. And recommending the product in the result to the client. Therefore, the customer portrait is comprehensively optimized according to the information of the target customer, and the recommended product is accurately matched.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence and the field of financial services, and in particular to a product recommendation method and device, equipment and a medium. BACKGROUND

[0002] In the increasingly competitive environment of the financial market, the traditional product recommendation method mainly relies on the basic information of customers, such as age, gender, income level, occupation, etc. Financial institutions simply classify customers according to these limited information, and then recommend the same financial products for each category. However, there are many drawbacks in practice, which is difficult to meet the needs of financial institutions and customers.

[0003] At present, the way of product recommendation is to use the historical transaction data of customers to recommend products. By analyzing the customer's past purchase records, transaction frequency, transaction amount and other information, the products that customers may be interested in are predicted. However, due to factors such as market environment and personal situation, the needs of customers will also change, and this product recommendation method is difficult to reflect the dynamic changes of customer demand in real time. For example, in the insurance product recommendation scenario, the customer is originally interested in high-risk investment insurance, but due to market changes or personal financial situation adjustment, he may prefer stable investment insurance products in the near future. However, due to the lack of timely information analysis, the insurance product still recommends high-risk investment insurance, resulting in the loss of the real needs of customers.

[0004] Therefore, how to optimize the customer portrait according to the information of the target customer and accurately match the recommended products has become a problem to be solved. SUMMARY

[0005] Therefore, the present application provides a product recommendation method, device, equipment and medium to solve the problem of how to optimize the customer portrait according to the information of the target customer and accurately match the recommended products.

[0006] In a first aspect, the present application provides a product recommendation method, comprising: After determining the target customer to be recommended, the customer basic information of the target customer is obtained, the target customer is portraited according to the customer basic information, the initial customer portrait is obtained, the initial customer portrait is matched with a preset product library to obtain a first matching result, and the product library stores the mapping relationship between the customer portrait and the product. if the first matching result contains a matched product, sending the matched product to the target customer and obtaining a first feedback result of the target customer on the matched product, and if the first feedback result is rejection, sending authorization verification information to the target customer, the authorization verification information being used to indicate whether the target customer authorizes to obtain corresponding customer social information; obtaining a second feedback result of the target customer on the authorization verification information, and if the second feedback result is permission, obtaining customer social information of the target customer, re-profiling the target customer according to the customer social information and the customer basic information, and obtaining an optimized customer profile; using the optimized customer profile to match the product library to obtain a second matching result, and sending products in the second matching result as recommended products to the target user.

[0007] In a second aspect, an embodiment of the present application provides a product recommendation device, which comprises: An initial matching module is configured to, after determining a target customer to be recommended, obtain customer basic information of the target customer, profile the target customer according to the customer basic information to obtain an initial customer profile, and use the initial customer profile to match a preset product library to obtain a first matching result, wherein the product library stores a mapping relationship between customer profiles and products. A verification feedback module is configured to, if the first matching result contains a matched product, send the matched product to the target customer and obtain a first feedback result of the target customer on the matched product, and if the first feedback result is rejection, send authorization verification information to the target customer, wherein the authorization verification information is used to indicate whether the target customer authorizes to obtain corresponding customer social information. A re-profiling module is configured to obtain a second feedback result of the target customer on the authorization verification information, and if the second feedback result is permission, obtain customer social information of the target customer, re-profile the target customer according to the customer social information and the customer basic information, and obtain an optimized customer profile. A recommendation module is configured to use the optimized customer profile to match the product library to obtain a second matching result, and send products in the second matching result as recommended products to the target user.

[0008] In a third aspect, an embodiment of the present application provides a computer device, which comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, and the processor implements the product recommendation method of the first aspect when executing the computer program.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the product recommendation method as described in the first aspect.

[0010] The beneficial effects of the embodiments in this application compared with the prior art are: In this application, after identifying the target customers to be recommended, the basic customer information of the target customers is obtained. Based on the basic customer information, a profile of the target customers is created to obtain an initial customer profile. The initial customer profile is then matched with a preset product library to obtain a first matching result. The product library stores the mapping relationship between customer profiles and products. If a matching product exists in the first matching result, the matching product is sent to the target customer, and the target customer's first feedback result on the matching product is obtained. If the first feedback result is rejection, authorization verification information is sent to the target customer. The authorization verification information is used to instruct the target customer to choose whether to authorize the acquisition of the corresponding customer social information. The target customer's second feedback result on the authorization verification information is obtained. If the second feedback result is permission, the target customer's customer social information is obtained. Based on the customer social information and the basic customer information, the target customer is re-profiled to obtain an optimized customer profile. The optimized customer profile is then matched with the product library to obtain a second matching result. The products in the second matching result are sent as recommended products to the target user. This application can be applied to insurance product recommendation scenarios. After identifying target customers, its basic information is obtained to construct an initial customer profile. This profile is then matched against a pre-defined product database to obtain a first matching result. If a matching product is found, it is sent to the customer. If the customer declines, an authorization verification message is sent. Once the customer grants authorization, their social information is obtained and combined with the basic information to create a more refined customer profile. This refined profile is then matched against the product database to obtain a second matching result. Products from this second matching result are then recommended to the customer. This comprehensive approach optimizes the customer profile based on the target customer's information, ensuring accurate product matching and recommendations. BRIEF DESCRIPTION OF DRAWINGS

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of an application environment for a product recommendation method provided in Embodiment 1 of this application; Figure 2 This is a flowchart illustrating a product recommendation method provided in Embodiment 2 of this application; Figure 3This is a flowchart illustrating a product recommendation method provided in Embodiment 3 of this application; Figure 4 This is a flowchart illustrating a product recommendation method provided in Embodiment 4 of this application; Figure 5 This is a flowchart illustrating a product recommendation method provided in Embodiment 5 of this application; Figure 6 This is a schematic diagram of the structure of a product recommendation device provided in Embodiment Six of this application; Figure 7 This is a schematic diagram of the structure of a computer device provided in Embodiment 7 of this application. DETAILED DESCRIPTION

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0015] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0016] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0017] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0018] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0019] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0020] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0021] It should be understood that the sequence number of each step in the following embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0022] To illustrate the technical solution of this application, specific embodiments are described below.

[0023] The product recommendation method provided in Embodiment 1 of this application can be applied to, for example, Figure 1 In this application environment, the client and server communicate with each other. Users can provide product recommendation conditions, requirements, and operation instructions through the client. The server then generates control instructions for the product recommendation method based on the content sent by the client. For example, in a financial system, a user performs a corresponding operation by manipulating the client. After receiving the operation, the server generates corresponding operation instructions and processes the operation instructions for the processor-cross-domain product recommendation method to obtain results containing product recommendations.

[0024] The client side includes, but is not limited to, PDAs, desktop computers, laptops, ultra-mobile personal computers (UMPCs), netbooks, cloud terminal devices, and personal digital assistants (PDAs). The server side can be a standalone server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0025] See Figure 2 This is a flowchart illustrating a product recommendation method provided in Embodiment 2 of this application. The above-described product recommendation method can be applied to... Figure 1 The server-side component.

[0026] like Figure 2 As shown, the product recommendation method may include the following steps: Step S201: After determining the target customer to be recommended, obtain the customer basic information of the target customer, create a profile of the target customer based on the customer basic information, obtain an initial customer profile, and match the initial customer profile with a preset product library to obtain a first matching result. The product library stores the mapping relationship between customer profiles and products.

[0027] Optionally, after matching the initial customer profile with a preset product database to obtain a first matching result, the following steps may also be included: Detect whether there is a matching product in the first matching result; If no matching product is found in the first matching result, then the step of sending authorization verification information to the target customer is executed.

[0028] There are several ways to identify target customers. For example, they can be screened based on the existing customer list of banks or financial institutions according to specific criteria (such as high-net-worth customers, new account opening customers, etc.). Alternatively, they can be targeted at specific groups (such as young office workers, retirees, etc.) based on the goals of the marketing campaign.

[0029] Basic customer information is the cornerstone of building customer profiles. This basic information includes personal information, financial information, occupational information, etc.

[0030] Customer profiling involves analyzing and integrating basic customer information to describe the customer using a series of tags and characteristics. For example, a 30-year-old, married customer with a monthly income of 10,000 yuan and working in a company might be profiled as a "middle-aged professional with stable income and family responsibilities, and certain financial management needs." An initial customer profile acts as a preliminary "template" for the customer, reflecting their basic characteristics and potential needs.

[0031] The pre-defined product database is a collection of products categorized and organized according to the needs and characteristics of different customer groups. This database stores the mapping between customer profiles and products; that is, different types of customer profiles correspond to different products. For example, the customer profile of a "young investor with high income and high risk tolerance" might correspond to products such as equity funds and private equity investments. Matching the initial customer profile with the product database involves identifying products that match that profile, thus obtaining the first matching result.

[0032] After obtaining the first matching results, it's necessary to check if any products are suitable for the target customer. This step is to determine the effectiveness of the products filtered based on the initial customer profile. If no matching products are found in the first matching results, it means that the initial profile built solely based on the customer's basic information cannot find suitable products in the existing product library. At this point, to further understand the customer's needs, an authorization verification message is sent to the target customer, inquiring whether they allow access to their corresponding social information. Customer social information may contain more information about the customer's interests, consumption habits, social circles, etc. This information helps build a more accurate customer profile, thereby finding products that are more suitable for the customer.

[0033] For example, in an insurance user information analysis scenario, an insurance company's financial product database stores mappings between different customer profiles and insurance products. Through matching, it was found that Mr. Li's initial customer profile matched a high-end auto insurance product from the company. This product offers high coverage amounts, comprehensive protection, and excellent claims service, making it suitable for customers like Mr. Li who have considerable financial resources and high-value vehicles. This yielded the first matching result.

[0034] Step S202: If there is a matching product in the first matching result, the matching product is sent to the target customer, and the first feedback result of the target customer on the matching product is obtained. If the first feedback result is a rejection, authorization verification information is sent to the target customer. The authorization verification information is used to instruct the target customer to choose whether to authorize the acquisition of the corresponding customer social information.

[0035] When a matching product is found in the first matching result, it means that a product that might be suitable for the target customer has been found in the product database based on the initial customer profile. At this point, the institution will push these matching products to the target customer. The push methods can be varied, including SMS, email, mobile banking app notifications, and phone calls with account managers. For example, a bank might use its mobile banking app to push information about wealth management products suitable for the customer's risk appetite and financial situation, including key details such as the product's returns, risk level, and investment period.

[0036] After sending the recommended products to the target customer, it's necessary to wait for their feedback. The initial feedback result mainly falls into two categories: acceptance and rejection. Customer acceptance means the initially recommended products meet their needs, while customer rejection indicates that these products fail to meet their expectations. Feedback can be obtained through customer actions on the software (such as clicking the accept or reject button), replying to text messages, or communicating with an account manager.

[0037] If the initial feedback is a refusal, it indicates that the initial customer profile built based on the customer's basic information is not accurate enough and fails to accurately reflect the customer's true needs. The authorization verification information is used to instruct the target customer whether to authorize the institution to access their corresponding social information. Customer social information contains a wealth of content, such as the customer's spending habits on social platforms (e.g., whether they frequently buy luxury goods or prioritize cost-effectiveness), interests (e.g., whether they enjoy travel, investment, etc.), and social circles (the occupations and spending levels of their friends). This information helps financial institutions gain a more comprehensive understanding of customers' lifestyles, values, and potential needs, thereby building more accurate customer profiles and recommending more suitable products.

[0038] For example, in an insurance customer information collection scenario, an insurance company recommends a high-end car insurance product to Mr. Li, informing him of its features and advantages. After consideration, Mr. Li rejects the product, possibly feeling the premium is too high or the coverage doesn't fully meet his needs. At this point, the insurance company sends Mr. Li authorization verification information, asking if he allows access to his social media information in order to recommend a more suitable insurance product.

[0039] Step S203: Obtain the second feedback result of the target customer on the authorization verification information. If the second feedback result is allowed, obtain the target customer's social information. Based on the customer's social information and the customer's basic information, re-profile the target customer to obtain an optimized customer profile.

[0040] When a target customer gives permission, it indicates that the customer is willing to allow the agency to access their social information, which provides richer material and possibilities for further understanding the customer.

[0041] Customer social information comes from a wide range of sources, including social media platform data, purchase records, and social relationship networks. Social media data can refer to a customer's activity level, the topics they follow, and the content they post on social media platforms such as WeChat, Weibo, and Douyin. For example, if a customer frequently follows financial investment topics, they may have a strong interest in and some understanding of financial products; if they frequently share travel-related content, they may have a need for travel-related consumer finance.

[0042] Consumer records refer to the data obtained through data integration with third-party payment platforms or partner merchants, which reveals customers' spending habits and preferences. For example, customers who frequently shop in high-end shopping malls may have high spending power and a pursuit of quality of life; those who frequently purchase fitness equipment may value a healthy lifestyle and have a need for related insurance or financial services.

[0043] Social networks refer to understanding a customer's social circle, including the professions and spending habits of their friends and family. If a customer's social circle consists mostly of entrepreneurs, then the customer may also have entrepreneurial aspirations or a potential need to participate in entrepreneurial-related investments.

[0044] After obtaining customer social information, it is integrated and analyzed with the previous customer basic information. By comprehensively considering factors such as customer interests, consumption habits, and social relationships reflected in the social information, as well as characteristics such as age, income, and occupation in the basic information, a new customer profile is created.

[0045] For example, in a financial product recommendation scenario, Mr. Li replied that he could access his social media information. The insurance company, through cooperation with social media platforms (within legal and compliant boundaries), obtained some information about Mr. Li on these platforms, discovering that he frequently participates in outdoor sports, is a member of a mountaineering enthusiast community, and often shares his travel experiences. Combining Mr. Li's basic information with this social media data, a new customer profile was created: "Middle-aged corporate middle manager, relatively high income, loves outdoor sports and travel, drives a luxury vehicle, and has a certain accident risk."

[0046] Step S204: Match the optimized customer profile with the product library to obtain a second matching result, and send the products in the second matching result as recommended products to the target user.

[0047] In the preceding steps, by integrating basic customer information and customer social information, we obtained an optimized customer profile. This profile more comprehensively and meticulously depicts the characteristics, needs, preferences, risk tolerance, and other aspects of the target user. For example, we might discover that the user not only has certain savings but is also enthusiastic about investing in emerging technology fields and has a relatively high risk tolerance.

[0048] The product database is a collection of various products and stores the mapping relationship between customer profiles and products. Different products correspond to different types of customer characteristics and needs. For example, high-risk, high-return equity funds may correspond to customer profiles with higher risk appetites and a desire for rapid asset growth, while stable bond funds are more suitable for customers with lower risk tolerance and a focus on preserving asset value.

[0049] Matching optimized customer profiles with a product database involves searching the database for the most suitable products based on the various characteristics and tags in the profile. This process may involve complex algorithms and data analysis to ensure accuracy. For example, the system might filter products that meet certain criteria based on factors such as the user's risk tolerance, investment goals, and financial situation, thus obtaining a second matching result.

[0050] After obtaining the second matching result, there will be multiple matched products. Financial institutions need to further screen and organize these products, considering factors such as profitability, risk, and liquidity, to provide users with the most valuable recommendations. For example, for a user with short-term idle funds, they might prioritize recommending highly liquid money market funds or short-term wealth management products.

[0051] There are several ways to send recommended products to target users, including SMS, email, mobile banking app push notifications, and phone calls with account managers. When pushing recommended products, detailed product information is provided, such as product name, expected return, risk level, and investment period, to help users make informed decisions. For example, when pushing recommended products to users through a mobile banking app, the product's features and advantages are presented in a visually appealing format with images and text.

[0052] For example, in a financial product recommendation scenario, the optimized customer profile is re-matched with the product database. It's discovered that, in addition to car insurance, Mr. Li is also suitable for a comprehensive accident insurance product that includes coverage for outdoor sports and travel. This product not only provides protection for his daily commutes but also offers corresponding risk protection during his outdoor activities and travel. The insurance company then sends this comprehensive accident insurance product to Mr. Li as a recommended product, completing the insurance product recommendation based on the optimized profile. In this application, after identifying the target customers to be recommended, the basic customer information of the target customers is obtained. Based on the basic customer information, a profile of the target customers is created to obtain an initial customer profile. The initial customer profile is then matched with a preset product library to obtain a first matching result. The product library stores the mapping relationship between customer profiles and products. If a matching product exists in the first matching result, the matching product is sent to the target customer, and the target customer's first feedback result on the matching product is obtained. If the first feedback result is rejection, authorization verification information is sent to the target customer. The authorization verification information is used to instruct the target customer to choose whether to authorize the acquisition of the corresponding customer social information. The target customer's second feedback result on the authorization verification information is obtained. If the second feedback result is permission, the target customer's customer social information is obtained. Based on the customer social information and the basic customer information, the target customer is re-profiled to obtain an optimized customer profile. The optimized customer profile is then matched with the product library to obtain a second matching result. The products in the second matching result are sent as recommended products to the target user. This application can be applied to insurance product recommendation scenarios. After identifying target customers, its basic information is obtained to construct an initial customer profile. This profile is then matched against a pre-defined product database to obtain a first matching result. If a matching product is found, it is sent to the customer. If the customer declines, an authorization verification message is sent. Once the customer grants authorization, their social information is obtained and combined with the basic information to create a more refined customer profile. This refined profile is then matched against the product database to obtain a second matching result. Products from this second matching result are then recommended to the customer. This comprehensive approach optimizes the customer profile based on the target customer's information, ensuring accurate product matching and recommendations.

[0053] See Figure 3 This is a flowchart illustrating a product recommendation method provided in Embodiment 3 of this application. Figure 3 As shown, in step S204 above, before sending the products in the second matching result as recommended products to the target user, the following steps may also be included: Step S301: Obtain the existing products of the target customer, extract the features of the existing products, and obtain the existing product features.

[0054] Step S302: Extract the features of each product in the second matching result to obtain the target product features of each product. For any product, compare the existing product features with the target product features corresponding to the product to obtain the comparison result.

[0055] Step S303: If the comparison result is the same or the similarity is greater than a preset value, then the product is removed from the second matching result to obtain an updated second matching result.

[0056] Step S304: Send the products in the updated second matching results as recommended products to the target user.

[0057] Institutions can use their own customer account systems to check the products currently held by target customers. These products may include savings accounts, various wealth management products, insurance products, funds, etc. For example, banks can view customers' account information with the bank to understand whether customers hold products such as time deposits and money market funds.

[0058] For each existing product acquired, its key characteristics need to be extracted. These key characteristics include product type (e.g., equity fund, bond fund), investment horizon (short-term, medium-term, long-term), return type (fixed income, variable income), risk level (low risk, medium risk, high risk), and investment sector (e.g., technology, consumer goods, finance). For example, for an equity fund, its key characteristics might be investing in the technology sector, medium-to-high risk, variable income, and long-term investment.

[0059] For each product in the second matching result, its key features are extracted to form the target product features. The extracted feature dimensions are consistent with the existing product features to ensure accurate comparison.

[0060] For each product in the second matching result, its target product features are compared one by one with the features of existing products. Various methods can be used for comparison, such as assigning feature weights, scoring the similarity of each feature, and then calculating the overall similarity. For example, important features such as product type and investment field can be assigned higher weights, while less important features can be assigned lower weights. Through this comparison, the similarity comparison results between each product and existing products are obtained.

[0061] The preset value is a similarity threshold set based on business needs and experience. For example, a similarity greater than 80% is set as a standard for high similarity.

[0062] If a product's comparison result with existing products is the same (i.e., features are completely identical) or the similarity is greater than a preset value, it means that the product is very similar to the customer's existing products, and recommending it to the customer again is not very meaningful. Therefore, the product is removed from the second matching results, resulting in an updated second matching result. This avoids recommending duplicate or highly similar products to the customer, improving the diversity and targeting of recommendations.

[0063] After the previous filtering and removal of similar products, the products in the updated second matching results are all products that are significantly different from the customer's existing products and are more novel and applicable.

[0064] The products from the updated second matching results are then used as the final recommended products and sent to the target users through appropriate channels (such as SMS, mobile banking app messages, email, etc.). This way, the recommended products received by users are more likely to meet their unmet financial needs, increasing user acceptance and satisfaction with the recommendations.

[0065] This application's embodiments optimize product recommendation results by comparing and filtering the features of existing products and matching products, providing more valuable recommendations for target users.

[0066] See Figure 4 This is a flowchart illustrating a product recommendation method provided in Embodiment 4 of this application. Figure 4 As shown, step S201, which involves creating a profile of the target customer based on the customer's basic information to obtain an initial customer profile, may include the following steps: Step S401: Extract customer age, customer gender, customer address, customer occupation, customer income and customer family structure from the customer basic information.

[0067] Step S402: Tag the target customer based on the customer's age, gender, address, occupation, income, and family structure to obtain a tag set as the initial customer profile of the target customer.

[0068] Optionally, the step of matching the initial customer profile with a preset product database to obtain a first matching result may include the following steps: The tags in the initial customer profile are matched one by one with the tags in the customer profiles in the preset product library to obtain the matching degree corresponding to each customer profile. If a customer profile exists that matches the preset conditions, then the product corresponding to the customer profile is determined to be the matching product in the first matching result. If no customer profile matches the preset conditions, the first matching result is determined to be empty.

[0069] Customer basic information is a series of fundamental data collected during the process of establishing relationships with customers. Key elements such as customer age, gender, address, occupation, income, and family structure are extracted from this information because they play a crucial role in understanding the customer's financial needs and risk tolerance.

[0070] Customers of different ages have different needs. For example, young people may be more concerned with consumer finance and small investments to meet their daily consumption and asset accumulation needs; while middle-aged customers may prefer stable investments and insurance planning to prepare for their families and future; and elderly customers are more focused on asset preservation and retirement security.

[0071] Customer gender can lead to differences in consumption and investment preferences between men and women. For example, women may place greater emphasis on the stability of their financial management, while men may be more willing to try higher-risk investments.

[0072] A customer's address can reflect information such as the economic development level and cost of living in their region. Customers living in economically developed areas may have higher spending power and more complex financial needs, while customers living in less developed areas may be more concerned with basic financial services.

[0073] A client's profession determines their income stability and the nature of their work. For example, civil servants and teachers have relatively stable incomes and may have a lower risk tolerance; while entrepreneurs and salespeople have more volatile incomes and may be more willing to take on certain risks for higher returns.

[0074] A customer's income level directly affects their spending power and investment scale. High-income customers may have more funds available for investment and wealth management, and have a greater demand for high-end products; while low-income customers are more focused on basic financial services and small savings.

[0075] A client's family structure includes whether they are married, have children, and the status of elderly care. For example, married clients with children need to consider their financial needs regarding their children's education and family security, while single clients may be more focused on personal consumption and investment.

[0076] Tagging involves transforming extracted customer information into tags with specific meanings to facilitate subsequent analysis and matching. Each key piece of information can correspond to one or more tags. For example, a customer who is 30 years old can be tagged as "youth". A customer whose profession is a doctor can be tagged as "medical industry" or "stable occupation". A customer with a high income can be tagged as "high-income group".

[0077] The tags in the initial customer profile are matched one-to-one with the tags in the customer profiles in the product library, and the degree of match is calculated. Each product in the product library has a pre-defined corresponding customer profile, which also exists in the form of tags. The tags in the initial customer profile of the target customer are compared with the tags of each customer profile in the product library, and the degree of match between them is calculated. The degree of match can be calculated using various methods, such as counting the number of identical tags or considering the importance weight of tags.

[0078] The preset condition is a threshold set by the organization based on business needs and experience. If the degree of matching between the customer profile corresponding to a certain product and the initial customer profile of the target customer reaches this threshold, it means that the product may be suitable for the target customer, and it is identified as the matching product in the first matching result.

[0079] If the matching degree between the customer profiles corresponding to all products and the initial customer profile of the target customer does not meet the preset conditions, it means that there are no suitable products for the target customer in the current product library. At this time, the first matching result is determined to be empty.

[0080] In this embodiment of the application, by constructing an initial customer profile and matching it with a product database, financial institutions can initially screen out products suitable for target customers, laying the foundation for subsequent accurate recommendations.

[0081] See Figure 5 This is a flowchart illustrating a product recommendation method provided in Embodiment 5 of this application. Figure 5 As shown, step S203, which involves re-profileing the target customer based on the customer's social information and basic customer information to obtain an optimized customer profile, may include the following steps: Step S501: Use a natural language model to analyze the text in the customer's social information to determine emotion-related keywords, lifestyle-related keywords, and first consumption description keywords.

[0082] Step S502: Using an image recognition model, analyze the images in the customer's social information to determine consumer-related images and work-related images. Perform feature description on the consumer-related images to obtain second consumer-related keywords, and perform feature description on the work-related images to obtain work-related keywords.

[0083] Step S503: Determine the health risk of the target customer based on the emotional keywords, the lifestyle keywords, and the customer basic information.

[0084] Step S504: Determine the occupational risk of the target customer based on the lifestyle keywords, the work keywords, and the customer basic information.

[0085] Step S505: Determine the financial risk of the target customer based on the first consumer description keywords, the second consumer description keywords, and the customer basic information.

[0086] Step S506: Based on the customer basic information, determine the family risk of the target customer, and re-profile the target customer based on the health risk, occupational risk, financial risk and family risk to obtain an optimized customer profile.

[0087] Optionally, the step of matching the optimized customer profile with the product database to obtain a second matching result may include the following steps: Perform feature analysis on all risks in the optimized customer profile to obtain customer demand feature results; Obtain the product feature results of feature analysis for each product in the product library; The customer demand characteristics are matched one by one with the product characteristics of each product to obtain a second matching result.

[0088] Among them, the text in customers' social information contains rich content, and natural language models can perform in-depth analysis of this text.

[0089] Emotional keywords are identified by analyzing text tone and word choice to discover words related to customer emotions, such as "happy," "anxious," and "frustrated." These keywords reflect the customer's current psychological state, which influences their decisions. For example, an anxious customer may be more inclined to choose stable products.

[0090] Lifestyle keywords include words that describe customers' lifestyles, interests, and hobbies, such as "travel," "fitness," and "reading." These words reflect customers' lifestyles and daily activities, helping to understand their needs and potential spending habits.

[0091] The first category of consumer description keywords involves expressions related to customer consumption, such as "purchasing luxury goods," "dining consumption," and "online shopping," which can directly reflect customers' consumption habits and consumption areas.

[0092] Image recognition models can automatically identify whether images in a customer's social media posts belong to a consumer or work-related scenario. Consumer-related images might include shopping receipts or restaurant photos, while work-related images could include office scenes or displays of work achievements.

[0093] By describing the features of consumer-related images, we can obtain second-level consumer-related keywords. For example, if the image is a photo of dining in a high-end restaurant, we might obtain keywords such as "high-end dining consumption." Similarly, by describing the features of work-related images, we can obtain work-related keywords. For example, if the image shows people working on a construction site, we could obtain keywords such as "construction industry work" or "outdoor work."

[0094] Health risks can be assessed by combining emotional keywords, lifestyle keywords, and basic customer information. For example, if a customer's basic information indicates an older age, emotional keywords suggest chronic anxiety, and lifestyle keywords indicate a lack of exercise, then it can be inferred that the customer may have a higher health risk, such as an increased risk of cardiovascular disease.

[0095] Occupational risks can be determined based on lifestyle keywords, work keywords, and basic customer information. For example, if a customer's basic information shows that they are a journalist, work keywords indicate that they frequently need to travel to dangerous areas for interviews, and lifestyle keywords indicate that they experience high work pressure and irregular work schedules, then it can be determined that the customer faces high occupational risks, such as personal safety risks and occupational health risks.

[0096] Financial risk is assessed by combining the primary and secondary consumer description keywords with basic customer information. If the basic customer information indicates average income, but the primary and secondary consumer description keywords both reflect a tendency towards high consumption, such as frequent purchases of luxury goods and upscale dining, then the customer may face significant financial risks, such as insufficient income to cover expenses or debt risks.

[0097] Family risk is determined based on factors such as family structure and economic status within the client's basic information. For example, if a client's family has multiple elderly members to support and relies on a single source of income, the family may face significant financial pressure and risks related to elder care.

[0098] By integrating health risks, occupational risks, financial risks, and family risks, we can re-profile target customers and obtain a more comprehensive, in-depth, and accurate optimized customer profile. This profile can more accurately reflect the customer's overall situation and potential needs.

[0099] A detailed analysis of health risks, occupational risks, financial risks, and family risks included in the optimized customer profile is conducted to extract key characteristics, such as the type, severity, and development trend of the risks, resulting in a profile of customer needs. This profile clearly demonstrates the characteristics of customers' financial needs when facing various risks.

[0100] Each product in the product database undergoes feature analysis, including its return characteristics, risk level, target audience, and coverage, resulting in product feature analysis for each product.

[0101] The customer's needs are compared one by one with the product characteristics of each product in the product database. Products with a high degree of matching with the customer's needs are identified as the second matching results. This matching method allows for the recommendation of products that better suit the customer's actual risk profile and financial needs.

[0102] Corresponding to the product recommendation method in the above embodiments, Figure 6 This paper shows a structural block diagram of a product recommendation device provided in Embodiment Six of this application. The product recommendation device can be applied to... Figure 1 The server-side component.

[0103] See Figure 6 The recommended device for this product includes: The initial matching module 61 is used to obtain the customer basic information of the target customer after determining the target customer to be recommended, to create a profile of the target customer based on the customer basic information, to obtain an initial customer profile, and to match the initial customer profile with a preset product library to obtain a first matching result. The product library stores the mapping relationship between customer profiles and products. The verification feedback module 62 is used to send the matching product to the target customer if there is a matching product in the first matching result, and obtain the first feedback result of the target customer on the matching product. If the first feedback result is a rejection, the module sends authorization verification information to the target customer. The authorization verification information is used to instruct the target customer to choose whether to authorize the acquisition of the corresponding customer social information. The re-profile module 63 is used to obtain the second feedback result of the target customer on the authorization verification information. If the second feedback result is allowed, the target customer's customer social information is obtained. Based on the customer social information and the customer basic information, the target customer is re-profiled to obtain an optimized customer profile. The recommendation module 64 is used to match the optimized customer profile with the product library to obtain a second matching result, and send the products in the second matching result as recommended products to the target user.

[0104] Optionally, the product recommendation device further includes: The matching product detection module is used to detect whether there are matching products in the first matching result after the initial customer profile is matched with the preset product library and a first matching result is obtained. The information sending module is used to send authorization verification information to the target customer if there is no matching product in the first matching result.

[0105] Optionally, the product recommendation device further includes: The existing product feature extraction module is used to obtain the existing products of the target customer and extract the features of the existing products before sending the products in the second matching result as recommended products to the target user. The similarity comparison module is used to extract the features of each product in the second matching result, obtain the target product features of each product, and compare the existing product features with the target product features corresponding to the product for any product to obtain the comparison result. The product removal module is used to remove the product from the second matching result if the comparison result is the same or the similarity is greater than a preset value, so as to obtain an updated second matching result; The product update module is used to send the products in the updated second matching results as recommended products to the target user.

[0106] Optionally, the initial matching module 61 includes: The information extraction unit is used to extract customer age, customer gender, customer address, customer occupation, customer income and customer family structure from the customer basic information; The tagging unit is used to tag the target customer based on the customer's age, gender, address, occupation, income, and family structure, so as to obtain a tag set as the initial customer profile of the target customer.

[0107] Optionally, the verification feedback module 62 includes: The tag matching unit is used to match the tags in the initial customer profile with the tags in the customer profiles in the preset product library one by one to obtain the matching degree corresponding to each customer profile. The matching product determination unit is used to determine the product corresponding to the customer profile as the matching product in the first matching result if there is a customer profile with a matching degree that meets the preset conditions. The matching result determination unit is used to determine that the first matching result is empty if there is no customer profile that matches the preset conditions.

[0108] Optionally, the re-image module 63 includes: The text analysis unit is used to analyze the text in the customer's social information using a natural language model to determine emotion-related keywords, lifestyle-related keywords, and primary consumption description keywords. The work-related keyword determination unit is used to analyze images in the customer's social information using an image recognition model, determine consumer-related images and work-related images, perform feature description on the consumer-related images to obtain second consumer-related keywords, and perform feature description on the work-related images to obtain work-related keywords. A health risk determination unit is used to determine the health risk of the target customer based on the emotional keywords, the lifestyle keywords, and the customer basic information. The occupational risk determination unit is used to determine the occupational risk of the target customer based on the lifestyle keywords, the work keywords, and the customer basic information. The financial risk determination unit is used to determine the financial risk of the target customer based on the first consumer description keywords, the second consumer description keywords, and the customer basic information. The re-profile unit is used to determine the family risk of the target customer based on the customer basic information, and to re-profile the target customer based on the health risk, occupational risk, financial risk and family risk to obtain an optimized customer profile.

[0109] Optionally, the recommendation module 64 includes: The risk analysis unit is used to perform feature analysis on all risks in the optimized customer profile to obtain customer demand feature results. The product feature analysis unit is used to obtain the product feature results of feature analysis for each product in the product library; The product feature result matching unit is used to match the customer demand feature result with the product feature result of each product one by one to obtain a second matching result.

[0110] It should be noted that the information interaction and execution process between the above modules, units, and sub-units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0111] Figure 7 This is a schematic diagram of the structure of a computer device provided in Embodiment Seven of this application. Figure 7 As shown, the computer device of this embodiment includes: at least one processor ( Figure 7 (Only one is shown in the diagram), a memory, and a computer program stored in the memory and executable on at least one processor, wherein the processor executes the computer program to implement the steps of any of the above-described product recommendation methods or product recommendation method embodiments.

[0112] This computer device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 7 The examples of computer devices are merely examples and do not constitute a limitation on computer devices. Computer devices may include more or fewer components than shown in the illustration, or combinations of certain components, or different components, such as network interfaces, displays, and input devices.

[0113] The processor referred to can be a CPU, but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0114] Memory includes readable storage media, internal memory, etc., wherein internal memory can be the RAM of a computer device, providing an environment for the operation of the operating system and computer-readable instructions stored in the readable storage media. The readable storage media can be the hard drive of a computer device, or in other embodiments, it can be an external storage device of the computer device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, memory can include both internal storage units and external storage devices of the computer device. Memory is used to store the operating system, applications, bootloader, data, and other programs, such as program code for computer programs. Memory can also be used to temporarily store data that has been output or will be output.

[0115] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here. If the integrated unit is implemented as 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, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium can include at least: any entity or device capable of carrying computer program code, a recording medium, a computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0116] The implementation of all or part of the processes in the methods of the above embodiments can also be accomplished by a computer program product. When the computer program product is run on a computer device, it enables the computer device to execute the steps in the above method embodiments.

[0117] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0118] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0119] In the embodiments provided in this application, it should be understood that the disclosed apparatus / computer devices and methods can be implemented in other ways. For example, the apparatus / computer device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0120] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0121] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A product recommendation method, characterized in that, include: After identifying the target customers to be recommended, the customer basic information of the target customers is obtained. Based on the customer basic information, the target customers are profiled to obtain an initial customer profile. The initial customer profile is then matched with a preset product library to obtain a first matching result. The product library stores the mapping relationship between customer profiles and products. If a matching product exists in the first matching result, the matching product is sent to the target customer, and the first feedback result of the target customer on the matching product is obtained. If the first feedback result is a rejection, authorization verification information is sent to the target customer. The authorization verification information is used to instruct the target customer to choose whether to authorize the acquisition of the corresponding customer social information. Obtain the second feedback result of the target customer on the authorization verification information. If the second feedback result is allowed, obtain the target customer's social information. Based on the customer's social information and the customer's basic information, re-profile the target customer to obtain an optimized customer profile. The optimized customer profile is matched with the product library to obtain a second matching result, and the products in the second matching result are sent as recommended products to the target user.

2. The product recommendation method according to claim 1, characterized in that, After matching the initial customer profile with a preset product database to obtain a first matching result, the process further includes: Detect whether there is a matching product in the first matching result; If no matching product is found in the first matching result, then the step of sending authorization verification information to the target customer is executed.

3. The product recommendation method according to claim 1, characterized in that, Before sending the products from the second matching result as recommended products to the target user, the method further includes: Obtain the existing products of the target customer, extract the features of the existing products, and obtain the existing product features; Extract the features of each product in the second matching result to obtain the target product features of each product. For any product, compare the existing product features with the target product features corresponding to the product to obtain the comparison result. If the comparison result is the same or the similarity is greater than a preset value, then the product is removed from the second matching result to obtain an updated second matching result; Sending the products from the second matching result as recommended products to the target user includes: The products in the updated second matching results are sent as recommended products to the target user.

4. The product recommendation method according to claim 1, characterized in that, The step of creating a profile of the target customer based on the customer's basic information to obtain an initial customer profile includes: The customer's age, gender, address, occupation, income, and family structure are extracted from the customer's basic information. Based on the customer's age, gender, address, occupation, income, and family structure, the target customer is tagged to obtain a tag set that forms the initial customer profile of the target customer.

5. The product recommendation method according to claim 4, characterized in that, The step of matching the initial customer profile with a preset product database to obtain a first matching result includes: The tags in the initial customer profile are matched one by one with the tags in the customer profiles in the preset product library to obtain the matching degree corresponding to each customer profile. If a customer profile exists that matches the preset conditions, then the product corresponding to the customer profile is determined to be the matching product in the first matching result. If no customer profile matches the preset conditions, the first matching result is determined to be empty.

6. The product recommendation method according to claim 1, characterized in that, The step of re-profileing the target customer based on the customer's social information and basic customer information to obtain an optimized customer profile includes: Using natural language models, the text in the customer's social information is analyzed to identify emotion-related keywords, lifestyle-related keywords, and primary consumption description keywords; Using an image recognition model, the images in the customer's social information are analyzed to determine consumer-related images and work-related images. The consumer-related images are characterized to obtain second consumer-related keywords, and the work-related images are characterized to obtain work-related keywords. Based on the emotional keywords, the lifestyle keywords, and the basic customer information, the health risks of the target customers are determined. Based on the lifestyle keywords, the work keywords, and the basic customer information, the occupational risks of the target customers are determined. Based on the first consumer description keywords, the second consumer description keywords, and the customer basic information, determine the financial risk of the target customer; Based on the customer's basic information, the family risk of the target customer is determined. Based on the health risk, occupational risk, financial risk, and family risk, the target customer is re-profiled to obtain an optimized customer profile.

7. The product recommendation method according to claim 6, characterized in that, The step of matching the optimized customer profile with the product database to obtain a second matching result includes: Perform feature analysis on all risks in the optimized customer profile to obtain customer demand feature results; Obtain the product feature results of feature analysis for each product in the product library; The customer demand characteristics are matched one by one with the product characteristics of each product to obtain a second matching result.

8. A product recommendation device, characterized in that, include: The initial matching module is used to obtain the customer basic information of the target customer after determining the target customer to be recommended, to create a profile of the target customer based on the customer basic information, to obtain an initial customer profile, and to match the initial customer profile with a preset product library to obtain a first matching result. The product library stores the mapping relationship between customer profiles and products. The verification feedback module is used to send the matching product to the target customer if there is a matching product in the first matching result, and obtain the first feedback result of the target customer on the matching product. If the first feedback result is a rejection, the module sends authorization verification information to the target customer. The authorization verification information is used to instruct the target customer to choose whether to authorize the acquisition of the corresponding customer social information. The re-profile module is used to obtain the second feedback result of the target customer on the authorization verification information. If the second feedback result is allowed, the target customer's social information is obtained. Based on the customer's social information and the customer's basic information, the target customer is re-profiled to obtain an optimized customer profile. The recommendation module is used to match the optimized customer profile with the product library to obtain a second matching result, and send the products in the second matching result as recommended products to the target user.

9. A computer device, characterized in that, The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the product recommendation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the product recommendation method as described in any one of claims 1 to 7.

Images