Product recommendation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a product recommendation method and device, electronic equipment and a storage medium, belongs to the technical field of product recommendation, and is suitable for the field of finance. The method comprises the following steps: acquiring alternative commercial record information of alternative commercial tenants; screening the alternative commercial tenants according to the alternative commercial record information to obtain target commercial tenants, and obtaining target commercial record information of the target commercial tenants; obtaining alternative insurance website information of preset alternative websites; according to the alternative insurance website information, the target business record information and the target merchant, screening the alternative websites to obtain a target website; performing insurance recommendation generation according to the target commercial record information to obtain a target insurance product; and performing insurance product recommendation on the target merchant by using the target insurance product according to the target website. The embodiment of the invention can improve the insurance product recommendation success rate of a website.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of product recommendation, and is suitable for the financial field, and particularly relates to a product recommendation method and device, an electronic device and a storage medium. BACKGROUND

[0002] Product recommendation refers to recommending products that a customer may be interested in according to the customer's needs. For example, an insurance outlet may recommend a series of insurance products to surrounding shops in a certain area. However, since the same area may contain multiple outlets and the customer types are diverse, the traditional recommendation method is often difficult to achieve efficient and accurate recommendation, thereby resulting in a low recommendation success rate. Therefore, how to improve the insurance product recommendation success rate of the outlet has become a problem to be solved. SUMMARY

[0003] The main purpose of the embodiments of the present application is to provide a product recommendation method and device, an electronic device and a storage medium, which aims to improve the insurance product recommendation success rate of the outlet.

[0004] To achieve the above-mentioned purpose, the first aspect of the embodiments of the present application provides a product recommendation method, which comprises:

[0005] obtaining candidate commercial record information of a candidate merchant;

[0006] screening the candidate merchant according to the candidate commercial record information to obtain a target merchant, and obtaining target commercial record information of the target merchant;

[0007] obtaining candidate insurance outlet information of a candidate outlet;

[0008] screening the candidate outlet according to the candidate insurance outlet information, the target commercial record information and the target merchant to obtain a target outlet;

[0009] generating insurance recommendation according to the target commercial record information to obtain a target insurance product;

[0010] recommending the target insurance product to the target merchant according to the target outlet.

[0011] In some embodiments, the screening of the candidate outlet according to the candidate insurance outlet information, the target commercial record information and the target merchant to obtain a target outlet comprises:

[0012] performing cooperation depth evaluation according to the candidate insurance outlet information and the target commercial record information to obtain cooperation depth data;

[0013] performing outlet address evaluation according to the candidate insurance outlet information to obtain address evaluation data;

[0014] The candidate locations are screened based on the cooperation depth data, the address evaluation data, and the target merchants to obtain the target locations.

[0015] In some embodiments, the step of filtering the candidate outlets based on the cooperation depth data, the address evaluation data, and the target merchant to obtain the target outlet includes:

[0016] Obtain the merchant address information of the target merchant and the branch address information of the alternative insurance outlets;

[0017] The distance between the merchant and the outlet is calculated based on the merchant address information and the outlet address information.

[0018] The distances between the merchants at the network outlets are normalized to obtain normalized distances;

[0019] The address evaluation data is weighted and fused based on the normalized distance to obtain weighted address data.

[0020] The candidate sites are filtered based on the weighted address data and the cooperation depth data to obtain the target sites.

[0021] In some embodiments, the step of filtering the candidate sites based on the weighted address data and the cooperation depth data to obtain the target site includes:

[0022] Based on the weighted address data and the cooperation depth data, a nine-square grid model is constructed to obtain the target nine-square grid model;

[0023] The sorting sequence of the candidate grid points is obtained from the nine-grid model to obtain the grid point sorting sequence;

[0024] The target network point is obtained by filtering the network point sorting sequence.

[0025] In some embodiments, the target business record information includes the target merchant's industry type, and the step of generating insurance recommendations based on the target business record information to obtain target insurance products includes:

[0026] Based on the target merchant's industry type, a preset insurance portfolio database is filtered to obtain a candidate insurance product portfolio;

[0027] The candidate insurance product combination and the target business record information are matched and evaluated according to the preset target product matching model to obtain matching evaluation data.

[0028] The candidate insurance product portfolio is screened based on the matching evaluation data to obtain the target insurance product.

[0029] In some embodiments, the target product matching model comprises an input embedding layer, a fusion layer, a flow network matching layer, and an output layer; the target business record information comprises merchant employee quantity information, merchant transaction history information, merchant claim record information, and merchant device record information; the matching evaluation of the target product matching model on the candidate insurance product combination and the target business record information comprises:

[0030] vector embedding of the candidate insurance product combination by the input embedding layer to obtain a product embedding vector, and vector embedding of the merchant employee quantity information, the merchant transaction history information, the merchant claim record information, the merchant device record information, and the target merchant industry type by the input embedding layer to obtain a merchant embedding vector;

[0031] vector fusion of the merchant embedding vector and the product embedding vector by the fusion layer to obtain a fusion embedding vector;

[0032] probability density calculation of the fusion embedding vector by the flow network matching layer to obtain a probability density distribution vector;

[0033] matching output of the probability density distribution vector by the output layer to obtain the matching evaluation data.

[0034] In some embodiments, the candidate business record information comprises a candidate merchant industry type, insurance purchase records, and business operation data; the screening of the candidate merchant according to the candidate business record information to obtain a target merchant comprises:

[0035] merchant industry risk evaluation according to the candidate merchant industry type to obtain merchant industry risk data;

[0036] customer stickiness analysis according to the insurance purchase records to obtain customer stickiness data;

[0037] operational efficiency evaluation according to the business operation data to obtain operational level data;

[0038] screening of the candidate merchant according to the merchant industry risk data, the customer stickiness data, and the operational level data to obtain the target merchant.

[0039] To achieve the above object, a second aspect of the embodiment of the present application proposes a product recommendation device, which comprises:

[0040] a first acquisition module configured to acquire candidate business record information of a candidate merchant;

[0041] The merchant filtering module is used to filter the candidate merchants based on the candidate business record information to obtain the target merchants and acquire the target business record information of the target merchants.

[0042] The second acquisition module is used to acquire information on potential insurance outlets from preset alternative outlets;

[0043] The outlet filtering module is used to filter the candidate outlets based on the candidate insurance outlet information, the target business record information, and the target merchant to obtain the target outlets;

[0044] The recommendation generation module is used to generate insurance recommendations based on the target business record information to obtain the target insurance product;

[0045] The insurance recommendation module is used to recommend insurance products to the target merchants based on the target outlets and the target insurance products.

[0046] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0047] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0048] This application proposes a product recommendation method, apparatus, electronic device, and storage medium. It obtains candidate business record information of potential merchants and, through screening these candidates, identifies target merchants. Further, by obtaining target business record information of the target merchants and preset candidate insurance outlet information, it screens the candidate outlets to obtain the most suitable target outlet. Subsequently, based on the target business record information, it generates target insurance products and recommends these products to the target merchants through the target outlets. Thus, this application's embodiments improve the success rate of insurance recommendations by screening and matching target merchants with target outlets. Attached Figure Description

[0049] Figure 1 This is a flowchart of the product recommendation method provided in the embodiments of this application;

[0050] Figure 2 yes Figure 1 The flowchart of step S102 in the document;

[0051] Figure 3 yes Figure 1 The flowchart of step S104 in the process;

[0052] Figure 4 is Figure 3 a flowchart of step S303 in

[0053] Figure 5 is Figure 4 a flowchart of step S405 in

[0054] Figure 6 is Figure 1 a flowchart of step S105 in

[0055] Figure 7 is Figure 6 a flowchart of step S602 in

[0056] Figure 8 is a structural schematic diagram of a product recommendation device provided by an embodiment of the present application;

[0057] Figure 9 is a hardware structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0059] It should be noted that although the functional modules are divided in the device schematic diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a manner different from the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification and claims and the above-described drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0061] Product recommendation refers to recommending products that may be of interest to customers according to the needs of customers. For example, in a certain area, insurance outlets will recommend a series of insurance products to surrounding shops. However, since the same area may contain multiple outlets and customers are of various types, traditional recommendation methods often have difficulty in achieving efficient and accurate recommendation, thereby resulting in a low success rate of recommendation. Therefore, how to improve the success rate of insurance product recommendation of outlets has become a problem to be solved.

[0062] Based on this, the product recommendation method and device, electronic equipment and storage medium provided in the embodiments of the present application aim to improve the success rate of insurance product recommendation of a network point.

[0063] The product recommendation method and device, electronic equipment and storage medium provided in the embodiments of the present application are specifically described through the following embodiments. First, the product recommendation method in the embodiments of the present application is described.

[0064] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Artificial intelligence (AI) is the theory, method, technology and application system for using a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results.

[0065] The basic technology of artificial intelligence generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. The software technology of artificial intelligence mainly includes computer vision technology, robot technology, biometric technology, speech processing technology, natural language processing technology and machine learning / deep learning, etc.

[0066] The product recommendation method provided in the embodiments of the present application relates to the technical field of product recommendation and is suitable for the financial field. The product recommendation method provided in the embodiments of the present application can be applied in a terminal, can be applied in a server end, and can also be software running in a terminal or a server end. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server end can be configured as an independent physical server, can be configured as a server cluster or a distributed system composed of multiple physical servers, can also be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN and big data and artificial intelligence platform, etc.; and the software can be an application for implementing the product recommendation method, but is not limited to the above forms.

[0067] The application is operable in a variety of general purpose or special purpose computer systems environments or configurations. Examples of well-known computing systems, environments, and / or configurations that can be suitable for use with the application include personal computers, server computers, handheld or laptop devices, tablet devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. The application can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. The application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules can be located in local and remote computer storage media including memory storage devices.

[0068] It should be noted that in each specific embodiment of the present application, when it is necessary to perform relevant processing according to user information, user behavior data, user history data, and user location information, and other data related to the identity or characteristics of the user, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the separate permission or separate consent of the user will be obtained through a pop-up window or by jumping to a confirmation page, and after obtaining the separate permission or separate consent of the user, the necessary user-related data for enabling the embodiments of the present application to function normally will be obtained.

[0069] Figure 1 is an optional flowchart of the product recommendation method provided by the embodiments of the present application, Figure 1 The method in the above embodiment can include, but is not limited to, steps S101 to S106.

[0070] Step S101, obtaining candidate commercial record information of a candidate merchant;

[0071] Step S102, screening the candidate merchant according to the candidate commercial record information to obtain a target merchant, and obtaining target commercial record information of the target merchant;

[0072] Step S103, obtaining candidate insurance site information of a preset candidate site;

[0073] Step S104, screening the candidate site according to the candidate insurance site information, the target commercial record information, and the target merchant to obtain a target site;

[0074] Step S105, generating an insurance recommendation according to the target commercial record information to obtain a target insurance product;

[0075] Step S106, recommending an insurance product to the target merchant based on the target insurance product and the target network point.

[0076] The steps S101 to S106 shown in the embodiments of the present application, by obtaining the alternative commercial record information of the alternative merchant, and by screening the alternative merchant, the target merchant is obtained. Further, by obtaining the target commercial record information of the target merchant and the preset alternative insurance network point information of the alternative network point, the alternative network point is screened to obtain the most suitable target network point. Then, based on the target commercial record information, the target insurance product is generated, and the insurance product is recommended to the target merchant through the target network point. In this way, the embodiments of the present application improve the success rate of insurance recommendation of the network point by screening and matching the target merchant and the target network point.

[0077] In step S101 of some embodiments, the alternative merchant refers to various merchants in a region, for example, various merchants in A province, for example, various merchants in A community. The alternative commercial record information includes merchant industry, insurance purchase record, commercial operation data, merchant employee quantity information, merchant transaction history record, merchant claim record information and merchant equipment record information, etc.

[0078] Different merchant industries have different needs for insurance. For example, manufacturing industry merchants usually need insurance products related to production equipment, worker safety, etc., while retail industry merchants may pay more attention to store, inventory, customer compensation, etc.

[0079] Insurance purchase record and merchant claim record information usually reflect the past insurance needs and insurance use of the merchant. The insurance purchase record of the merchant can help to judge its preference for insurance products and purchase frequency, and the claim record information can reflect the risk situation of the merchant in the past insurance use. If the merchant has a higher claim record, those products with wider protection range and lower risk can be recommended, and if the claim record of the merchant is less, the basic insurance product can be recommended.

[0080] Commercial operation data and merchant transaction history record can determine the operating status and market performance of the merchant. For example, the transaction history record of the merchant reflects the business activity, sales, customer group, etc. These data are helpful to assess the operating risk of the merchant and recommend suitable insurance products. The commercial operation data, such as turnover, sales mode, product category, etc., can further refine the insurance product recommendation.

[0081] The number of employees of a merchant determines the type and protection limit of the insurance it needs to a certain extent. For example, a merchant with a large number of employees may pay more attention to work injury, medical insurance, etc.

[0082] The business equipment record information includes the conditions of the production equipment, office facilities and other assets owned by the business. According to the data such as the type of equipment, service life, maintenance condition and the like, the type of equipment protection that the business may need can be determined, such as loss insurance of production equipment, equipment failure insurance and the like.

[0083] Please refer to Figure 2 In some embodiments, the alternative business record information includes an alternative business industry type, insurance purchase record and business operation data, and step S102 can include but is not limited to steps S201 to S204:

[0084] Step S201, performing business industry risk assessment according to the alternative business industry type to obtain business industry risk data;

[0085] Step S202, performing customer stickiness analysis according to the insurance purchase record to obtain customer stickiness data;

[0086] Step S203, performing operation efficiency evaluation according to the business operation data to obtain operation level data;

[0087] Step S204, screening the alternative business according to the business industry risk data, the customer stickiness data and the operation level data to obtain a target business.

[0088] The steps S201 to S204 shown in the embodiments of the present application perform business industry risk assessment on the industry type of the alternative business to obtain business industry risk data, and perform customer stickiness analysis on the insurance purchase record to obtain customer stickiness data. Meanwhile, the embodiments of the present application also perform operation efficiency evaluation on the business operation data to obtain operation level data. Then, the embodiments of the present application perform comprehensive analysis on the business industry risk data, the customer stickiness data and the operation level data to screen the alternative business and obtain a target business. In this way, the embodiments of the present application accurately find the target business that may need insurance recommendation through comprehensive evaluation of business industry risk, customer stickiness and operation efficiency.

[0089] In step S201 of some embodiments, the business industry risk assessment refers to the assessment according to the industry type of the alternative business, specifically, generating corresponding business industry risk data by analyzing the historical data and market trends of the industry. The business industry risk data is a series of levels, such as high risk and low risk.

[0090] For example, the historical data corresponding to the business industry type is obtained from a preset database. For example, for the automobile manufacturing industry, the profit is rising, and it can be considered as low risk.

[0091] In step S202 of some embodiments, the customer stickiness analysis refers to analysis according to the insurance purchase records of the alternative business, specifically, by analyzing the historical purchase behavior, purchase frequency, purchase amount, renewal rate and other data of the customer, corresponding customer stickiness data is generated. The customer stickiness data can reflect the customer's loyalty and purchase potential of the insurance product, for example, high stickiness customers may be high-value customers.

[0092] For example, the purchase history data of the customer is obtained from the preset database, for example, if a customer will purchase the same type of insurance product every year, and the purchase amount increases year by year, it can be considered that the customer has high stickiness, and belongs to a high stickiness customer.

[0093] In step S203 of some embodiments, the operating efficiency evaluation refers to evaluation according to the business operation data of the alternative business, specifically, by analyzing the turnover, sales mode, product category and other data of the business, corresponding operating level data is generated. The operating level data can reflect the operating ability and market performance of the business, and is usually used to evaluate the profit potential and stability of the business.

[0094] For example, the operating history data of the business is obtained from the preset database, for example, the annual turnover of a certain retail business continues to grow, and the sales products have high market demand, so it can be considered that the business has high operating efficiency, and belongs to a high-level business.

[0095] In step S204 of some embodiments, the target business is obtained by screening the alternative business according to the business industry risk data, customer stickiness data and operating level data. Specifically, the business industry risk data, customer stickiness data and operating level data correspond to an evaluation index, and when the value of the evaluation index is greater than a preset threshold, the alternative business is determined as the target business.

[0096] In step S103 of some embodiments, the preset alternative insurance point information of the alternative point is obtained, wherein the alternative point refers to each insurance point in a specific area, for example, the insurance points in A province or the insurance points within A community. The alternative insurance point information includes related data of the point, such as the operation of the point, the type of the sales product, the customer group of the point, the coverage of the insurance product, the sales history record, the conversion rate, etc.

[0097] Please refer to Figure 3 In some embodiments, step S104 can include but is not limited to steps S301 to S303:

[0098] Step S301, cooperation depth evaluation is performed according to the alternative insurance point information and the target business record information, and cooperation depth data is obtained;

[0099] Step S302, according to the alternative insurance site information, the site address is evaluated, and address evaluation data is obtained;

[0100] Step S303, according to the cooperation depth data, the address evaluation data and the target merchant, the alternative site is screened, and the target site is obtained.

[0101] The steps S301 to S303 shown in the embodiments of the application are that the cooperation depth data is obtained by evaluating the cooperation depth of the alternative insurance site according to the alternative insurance site information and the target business record information; and the address evaluation data is obtained by evaluating the address of the site according to the alternative insurance site information. Then, the target site is obtained by screening the alternative site according to the cooperation depth data, the address evaluation data and the target merchant. In this way, the target site suitable for recommending insurance to the target merchant is accurately screened by comprehensively evaluating the cooperation depth data and the address evaluation data.

[0102] In step S301 of some embodiments, the cooperation depth evaluation refers to evaluating the alternative insurance site according to the alternative insurance site information and the target business record information, specifically, by analyzing the historical cooperation situation, cooperation frequency, cooperation scale of the site and the business behavior data of the merchant, the corresponding cooperation depth data is generated. The cooperation depth data can represent the closeness of cooperation between the merchant and the insurance site.

[0103] In step S302 of some embodiments, the site address evaluation refers to evaluating the geographical location of the site according to the alternative insurance site information, specifically, by analyzing the geographical location of the alternative insurance site and the distribution of the surrounding merchants through the geographic information system (GIS), the radiation business opportunity density of the site is evaluated. Through the evaluation, the corresponding address evaluation data is generated, reflecting the market potential and radiation ability of the region where the site is located.

[0104] Please refer to Figure 4 In some embodiments, step S303 can include but is not limited to steps S401 to S405:

[0105] Step S401, obtaining the merchant address information of the target merchant, and obtaining the site address information of the alternative insurance site information;

[0106] Step S402, according to the merchant address information and the site address information, the distance is calculated, and the site-merchant distance is obtained;

[0107] Step S403, the site-merchant distance is normalized, and the normalized distance is obtained;

[0108] Step S404, according to the normalized distance, the address evaluation data is weighted and fused, and the weighted address data is obtained;

[0109] Step S405: screening the candidate insurance points according to the weighted address data and the cooperation depth data to obtain the target insurance point.

[0110] The steps S401-S405 shown in the embodiments of the present application are as follows: the merchant address information of the target merchant is obtained, and the address information of the candidate insurance point is obtained; the distance between the merchant and the insurance point is calculated according to the merchant address information and the address information of the insurance point to obtain the distance between the merchant and the insurance point; then, the distance between the merchant and the insurance point is normalized to obtain the normalized distance; the address evaluation data is weighted and fused according to the normalized distance to obtain the weighted address data; finally, the candidate insurance point is screened according to the weighted address data and the cooperation depth data to obtain the target insurance point. In this way, the distance between the merchant and the candidate insurance point is calculated, and the address evaluation data and the cooperation depth data are weighted and fused, so that the target insurance point is accurately screened.

[0111] In step S401 of some embodiments, the merchant address information refers to the geographic position data of the target merchant, specifically the latitude and longitude. The address information of the insurance point refers to the geographic position data of the candidate insurance point, specifically the latitude and longitude and the like.

[0112] In step S402 of some embodiments, the distance calculation according to the merchant address information and the address information of the insurance point is shown in formula (1):

[0113]

[0114] wherein d is the physical distance between the target merchant and the candidate insurance point, r is the radius of the earth, Δφ = φ2- φ1 is the difference in longitude between the target merchant and the target insurance point, Δλ = λ2- λ1 is the difference in latitude between the target merchant and the target insurance point, φ1, φ2 are the latitudes of the target merchant and the candidate insurance point respectively, and λ1, λ2 are the longitudes of the target merchant and the candidate insurance point respectively.

[0115] In step S403 of some embodiments, the normalization processing refers to performing proportional operation on the distance between the merchant and the insurance point according to the maximum value of the region where the target merchant and the target insurance point are located as the reference value to obtain the normalized distance.

[0116] In step S404 of some embodiments, the product of the normalized distance and the address evaluation data is obtained to obtain weighted address data. It should be noted that the product of the normalized distance and the address evaluation data is obtained because the address evaluation data represents the market potential, service capacity and coverage of the insurance outlets in the target region, and the normalized distance represents the spatial distance between the target business and the candidate outlet. The combination of the two can optimize the outlet selection based on the geographical location factor, ensure that the outlet with a closer distance and greater market potential is more likely to be selected, and thus improve the accuracy and effect of the screening process.

[0117] Referring to Figure 5 In some embodiments, step S405 includes but is not limited to steps S501 to S503:

[0118] In step S501, a nine-square model is constructed according to the weighted address data and the cooperation depth data to obtain a target nine-square model.

[0119] In step S502, a sorting sequence of the candidate outlet is obtained from the nine-square model to obtain an outlet sorting sequence.

[0120] In step S503, the outlet sorting sequence is screened to obtain a target outlet.

[0121] The steps S501 to S503 shown in the embodiments of the present application construct a nine-square model according to the weighted address data and the cooperation depth data to obtain a target nine-square model, then obtain a sorting sequence of the candidate outlet from the target nine-square model to obtain an outlet sorting sequence, and finally screen the outlet sorting sequence to obtain a target outlet. In this way, the embodiments of the present application fuse the weighted address data and the cooperation depth data into the nine-square model, and select the most suitable target outlet through the sorting of the nine-square model.

[0122] In step S501 of some embodiments, the nine-square model construction refers to model construction according to the weighted address data and the cooperation depth data, specifically classifying the candidate outlets according to the cooperation depth and the radiation business opportunity density to generate a target nine-square model. For example, if the weighted address data is high level and the cooperation depth data is high level, it is a level I outlet of the nine-square model; if the weighted address data is high level and the cooperation depth data is medium level, it is a level II outlet of the nine-square model; if the weighted address data is medium level and the cooperation depth data is low level, it is a level IV outlet of the nine-square model; and if the weighted address data is low level and the cooperation depth data is low level, it is a level IX outlet of the nine-square model.

[0123] In step S502 of some embodiments, all candidate grid points are sorted according to their levels in the nine-square grid model according to the priority of the grid points in the nine-square grid model, to obtain a sorting sequence of the grid points.

[0124] In step S503 of some embodiments, the grid point sorting sequence is filtered, that is, the grid point corresponding to the highest level in the nine-square grid model is obtained, to obtain a target grid point.

[0125] Referring to Figure 6 In some embodiments, the target business record information includes a target business industry type, and step S105 includes but is not limited to steps S601 to S603.

[0126] In step S601, the target business industry type is used to filter the preset insurance portfolio database, to obtain a candidate insurance product portfolio.

[0127] In step S602, the candidate insurance product portfolio and the target business record information are matched and evaluated according to a preset target product matching model, to obtain matching evaluation data.

[0128] In step S603, the candidate insurance product portfolio is filtered according to the matching evaluation data, to obtain a target insurance product.

[0129] The steps S601 to S603 shown in the embodiments of the present application filter the preset insurance portfolio database according to the target business industry type to obtain a candidate insurance product portfolio. Then, the system matches and evaluates the candidate insurance product portfolio and the target business record information according to a preset target product matching model to obtain matching evaluation data. Finally, the candidate insurance product portfolio is filtered according to the matching evaluation data to obtain a target insurance product. In this way, the embodiments of the present application match and evaluate the candidate insurance product portfolio and the target business industry type through the target product matching model, to accurately obtain a target insurance product suitable for the target business.

[0130] In step S601 of some embodiments, the preset insurance portfolio database is a database storing different types of insurance product portfolios, wherein each product portfolio is designed according to industry characteristics and business needs. For example, a business in the manufacturing industry usually needs insurance products covering production equipment, worker safety, and work injury compensation. A business in the retail industry is more concerned about store property insurance, inventory insurance, and customer compensation.

[0131] Referring to Figure 7In some embodiments, the target product matching model comprises an input embedding layer, a fusion layer, a flow network matching layer, and an output layer; the target business record information comprises merchant employee quantity information, merchant transaction history information, merchant claim record information, and merchant device record information; and step S602 can comprise, but is not limited to, steps S701-S704.

[0132] In step S701, the input embedding layer is used to perform vector embedding on the candidate insurance product combination to obtain a product embedding vector, and the input embedding layer is used to perform vector embedding on the merchant employee quantity information, the merchant transaction history information, the merchant claim record information, the merchant device record information, and the target merchant industry type to obtain a merchant embedding vector.

[0133] In step S702, the fusion layer is used to perform vector fusion on the merchant embedding vector and the product embedding vector to obtain a fusion embedding vector.

[0134] In step S703, the flow network matching layer is used to perform probability density calculation on the fusion embedding vector to obtain a probability density distribution vector.

[0135] In step S704, the output layer is used to perform matching output on the probability density distribution vector to obtain matching evaluation data.

[0136] The steps S701-S704 shown in the embodiments of the present application perform vector embedding on the candidate insurance product combination and the merchant employee quantity information, the merchant transaction history information, the merchant claim record information, the merchant device record information, and the target merchant industry type through the input embedding layer to obtain the product embedding vector and the merchant embedding vector. Then, the fusion layer is used to perform vector fusion on the merchant embedding vector and the product embedding vector to obtain the fusion embedding vector. Then, the flow network matching layer is used to perform probability density calculation on the fused embedding vector to obtain the probability density distribution vector, and finally the output layer is used to perform matching output on the probability density distribution vector to obtain the matching evaluation data. In this way, the embodiments of the present application can accurately evaluate the most suitable insurance product combination for the target merchant through vector embedding, fusion, and flow network matching of multi-dimensional data.

[0137] In steps S701-S704 of some embodiments, the target product matching model is composed of an input embedding layer, a flow network matching layer, and an output layer.

[0138] The input embedding layer performs input embedding through a multi-layer perception (MLP) and a gating module. Specifically, the multi-layer perception (MLP) is used to extract features of the candidate insurance product combination and the merchant information, and convert them into embedding vectors suitable for subsequent processing. The gating module is used to control the selective transmission of information flow, further enhancing the expression ability of the model.

[0139] The flow network matching layer is formed by a flow network matching module, which is used for matching calculation according to the fusion embedding vector of the merchant embedding vector and the product embedding vector. The module calculates the matching degree between the merchant and the insurance product according to the input feature vector, and generates a probability density distribution vector of the matching.

[0140] The output layer is composed of an activation function module, which is mainly used for outputting the matching evaluation data. Specifically, the output layer normalizes the matching evaluation data to a value between 0 and 1 through a sigmoid activation function, representing the matching probability of the merchant and the insurance product combination, and finally determining which insurance product combination is the most matched with the merchant.

[0141] It should be noted that the flow network has a high accuracy in the matching of insurance product combinations, because the probability calculation characteristics of the flow network can effectively handle the matching relationship between the merchant and the insurance product combination. Unlike traditional matching methods, the flow network simulates the flow by taking the merchant features and product features as different nodes, and calculates the probability density of matching according to the connection relationship between these nodes, thereby quantifying the adaptation degree between the merchant and the insurance product combination.

[0142] In the embodiments of the present application, the relationship between the merchant and the insurance product combination is not a simple binary matching (i.e. matching or not matching), but a probability distribution problem involving multiple possibilities. The flow network can perform probability calculation on the matching of each product combination according to multiple input features (such as the industry type, transaction history, and claim record of the merchant), and generate a probability density distribution vector, thereby evaluating the adaptation degree of each alternative product combination under the demand of the merchant.

[0143] The probabilistic output of the flow network is suitable for handling complex matching problems, because it can effectively consider the different influences of each feature on the matching, avoiding the limitations brought by hard decisions (i.e. a certain insurance in the insurance product combination is not a binary matching for a certain feature of the merchant).

[0144] In step S603 of some embodiments, the alternative insurance product combinations are screened according to the matching evaluation data, that is, the alternative insurance product combination with the highest probability is selected, and the target insurance product is obtained.

[0145] In step S106 of some embodiments, after the target insurance product is determined, the target insurance product will be handed over to the target network point for product recommendation to the target merchant. The specific recommendation methods can include short message recommendation, on-site recommendation by a business staff, telephone recommendation, etc.

[0146] Please refer to Figure 8 The embodiments of the present application also provide a product recommendation device, which can implement the above product recommendation method. The device comprises:

[0147] The first obtaining module 801 is configured to obtain candidate business record information of a candidate merchant;

[0148] The merchant screening module 802 is configured to screen the candidate merchant according to the candidate business record information, to obtain a target merchant, and to obtain target business record information of the target merchant;

[0149] The second obtaining module 803 is configured to obtain candidate insurance site information of a preset candidate site;

[0150] The site screening module 804 is configured to screen the candidate site according to the candidate insurance site information, the target business record information, and the target merchant, to obtain a target site;

[0151] The recommendation generation module 805 is configured to generate an insurance recommendation according to the target business record information, to obtain a target insurance product;

[0152] The insurance recommendation module 806 is configured to recommend the target insurance product to the target merchant according to the target site and the target insurance product.

[0153] The specific implementation of the product recommendation apparatus is basically the same as that of the above-mentioned product recommendation method, and thus will not be repeated here.

[0154] The embodiments of the present application further provide an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the above-mentioned product recommendation method when executing the computer program. The electronic device can be any intelligent terminal, such as a tablet computer or a vehicle-mounted computer.

[0155] Please refer to Figure 9 , Figure 9 The hardware structure of the electronic device of another embodiment is illustrated, which includes:

[0156] The processor 901 can be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, and is configured to execute a related program to implement the technical solutions provided by the embodiments of the present application.

[0157] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 902 can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the related program codes are stored in the memory 902 and are called and executed by the processor 901 to implement the product recommendation method of the embodiments of the present application;

[0158] The input / output interface 903 is configured to realize information input and output.

[0159] The communication interface 904 is configured to realize the communication interaction between the device and other devices. The communication can be realized by a wired manner (for example, a USB, a network cable, etc.) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, etc.).

[0160] The bus 905 is configured to transmit information between various components (for example, the processor 901, the memory 902, the input / output interface 903, and the communication interface 904) of the device.

[0161] The processor 901, the memory 902, the input / output interface 903, and the communication interface 904 are connected to each other through the bus 905 to realize the communication connection between the device.

[0162] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the product recommendation method.

[0163] The memory is a non-transitory computer readable storage medium, which can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor. These remote memories can be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0164] The product recommendation method, product recommendation device, electronic equipment and storage medium provided by the embodiments of the present application obtain the candidate commercial record information of the candidate merchant, and obtain the target merchant by screening the candidate merchant. Further, the target commercial record information of the target merchant and the candidate insurance site information of the preset candidate site are obtained, the candidate site is screened to obtain the most suitable target site. Then, the target insurance product is generated based on the target commercial record information, and the insurance product is recommended to the target merchant through the target site. In this way, the embodiments of the present application improve the success rate of insurance recommendation of the site by screening and matching the target merchant and the target site.

[0165] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0166] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and can include more or fewer steps than the figures shown, or combine certain steps, or different steps.

[0167] The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separate, that is, can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0168] Those skilled in the art can understand that all or some of the steps in the above disclosed method, the function modules / units in the system and the device can be implemented as software, firmware, hardware and their appropriate combinations.

[0169] The terms "first", "second", "third", "fourth" and the like (if any) in the specification of the present application and the above-described figures are used to distinguish similar objects, and do not necessarily have to describe a particular order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, 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 "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0170] It should be understood that, in the application, "at least one" refers to one or more, and "multiple" refers to two or more. "And / or" is used to describe the association relationship of the associated objects, which means that there can be three relationships, for example, "A and / or B" can represent three cases of only A, only B, and A and B existing at the same time, wherein A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after it. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can represent a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, and c can be single or multiple.

[0171] In several embodiments provided in the application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the above units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed mutual ones can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0172] The units described above as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0173] In addition, each functional unit in each embodiment of the application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0174] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.

[0175] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, and are not limited to the scope of the embodiments of the present application. Any modifications, equivalent replacements and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the embodiments of the present application.

Claims

1. A product recommendation method characterized by, The method comprises: obtaining candidate commercial record information of a candidate merchant; screening the candidate merchant according to the candidate commercial record information to obtain a target merchant, and obtaining target commercial record information of the target merchant; obtaining candidate insurance site information of a candidate site; screening the candidate site according to the candidate insurance site information, the target commercial record information and the target merchant to obtain a target site; generating an insurance recommendation according to the target commercial record information to obtain a target insurance product; recommending the target insurance product to the target merchant according to the target site.

2. The method of claim 1, wherein, The screening of the candidate site according to the candidate insurance site information, the target commercial record information and the target merchant to obtain a target site comprises: performing cooperation depth evaluation according to the candidate insurance site information and the target commercial record information to obtain cooperation depth data; performing site address evaluation according to the candidate insurance site information to obtain address evaluation data; screening the candidate site according to the cooperation depth data, the address evaluation data and the target merchant to obtain the target site.

3. The method of claim 2, wherein, The screening of the candidate site according to the cooperation depth data, the address evaluation data and the target merchant to obtain the target site comprises: obtaining merchant address information of the target merchant and obtaining site address information of the candidate insurance site information; performing distance calculation according to the merchant address information and the site address information to obtain site-merchant distance; performing normalization processing on the site-merchant distance to obtain normalized distance; performing weighted fusion on the address evaluation data according to the normalized distance to obtain weighted address data; screening the candidate site according to the weighted address data and the cooperation depth data to obtain the target site.

4. The method of claim 3, wherein, The screening of the candidate site according to the weighted address data and the cooperation depth data to obtain the target site comprises: performing nine-square grid model construction according to the weighted address data and the cooperation depth data to obtain a target nine-square grid model; obtaining a sorting sequence of the candidate site from the nine-square grid model to obtain a site sorting sequence; screening the site sorting sequence to obtain the target site.

5. The method of claim 1, wherein, The target commercial record information comprises a target merchant industry type, and the generation of a target insurance product according to the target commercial record information comprises: screening a preset insurance combination database according to the target merchant industry type to obtain a candidate insurance product combination; performing matching evaluation on the candidate insurance product combination and the target commercial record information according to a preset target product matching model to obtain matching evaluation data; screening the candidate insurance product combination according to the matching evaluation data to obtain the target insurance product.

6. The method of claim 5, wherein, The target product matching model comprises an input embedding layer, a fusion layer, a flow network matching layer and an output layer; the target business record information comprises merchant employee quantity information, merchant transaction history information, merchant claim record information and merchant equipment record information; the matching evaluation data obtained by matching and evaluating the candidate insurance product combination and the target business record information according to the preset target product matching model comprises: the product embedding vector obtained by performing vector embedding on the candidate insurance product combination through the input embedding layer, and the merchant embedding vector obtained by performing vector embedding on the merchant employee quantity information, the merchant transaction history information, the merchant claim record information, the merchant equipment record information and the target merchant industry type through the input embedding layer; the fusion embedding vector obtained by performing vector fusion on the merchant embedding vector and the product embedding vector through the fusion layer; the probability density distribution vector obtained by performing probability density calculation on the fusion embedding vector through the flow network matching layer; the matching evaluation data obtained by performing matching output on the probability density distribution vector through the output layer.

7. The method of claim 1, wherein, The candidate business record information comprises a candidate merchant industry type, insurance purchase records and business operation data; the target merchant obtained by screening the candidate merchant according to the candidate business record information comprises: merchant industry risk data obtained by performing merchant industry risk evaluation according to the candidate merchant industry type; customer stickiness data obtained by performing customer stickiness analysis according to the insurance purchase records; operating level data obtained by performing operating efficiency evaluation according to the business operation data; the target merchant obtained by screening the candidate merchant according to the merchant industry risk data, the customer stickiness data and the operating level data.

8. A product recommendation device characterized by comprising: The device comprises: a first acquisition module configured to acquire candidate business record information of a candidate merchant; a merchant screening module configured to screen the candidate merchant according to the candidate business record information to obtain a target merchant and acquire target business record information of the target merchant; a second acquisition module configured to acquire candidate insurance site information of a preset candidate site; a site screening module configured to screen the candidate site according to the candidate insurance site information, the target business record information and the target merchant to obtain a target site; a recommendation generation module configured to generate an insurance recommendation according to the target business record information to obtain a target insurance product; an insurance recommendation module configured to recommend the target insurance product to the target merchant according to the target site.

9. An electronic device, comprising: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the product recommendation method in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the product recommendation method in any one of claims 1 to 7.

Images