CN118886986A - Product recommendation method, device, equipment and storage medium - Google Patents

Abstract

The application discloses a product recommendation method, a device, equipment and a storage medium, and relates to the technical field of information processing, wherein the product recommendation method comprises the following steps: obtaining a product recommendation request of a target guest group, wherein the product recommendation request comprises a customer portrait; determining a product recommendation list corresponding to the target guest group based on the guest representation; configuring a product recommendation strategy based on the product recommendation request and the product recommendation list; and recommending the products to the target guest group based on the product recommendation strategy. According to the method, the product recommendation strategy corresponding to the client image of the target customer group is generated, and personalized product recommendation is provided for the actual demand and purchase preference of each client, so that the accuracy and success rate of product recommendation are improved, the market competitiveness of the product is effectively improved, and the satisfaction degree of the client is improved.

Description

Product recommendation method, device, equipment and storage medium

Technical Field

The present application relates to the field of information processing technology, and in particular to a product recommendation method, device, equipment and storage medium.

Background Art

Most of the existing product recommendation solutions collect the number of product clicks by customers on mobile phones, online banking and other channels, and then identify the customer's purchase intention through the number of visits, and then make product recommendations. However, although such solutions can make product recommendations to customers, the recommendation strategy only analyzes customer visit behavior and cannot accurately capture the customer's actual needs and purchase intentions, resulting in a deviation between the recommended products and the customer's real needs, which reduces the customer's trust in the recommendation system and causes customers to miss out on the financial products that are truly suitable for them. At the same time, it also leads to a waste of promotion resources for financial products, which in the long run leads to customer loss and makes financial products lose their market competitiveness.

Summary of the invention

The main purpose of this application is to provide a product recommendation method, device, equipment and storage medium, aiming to improve the accuracy and success rate of product recommendations, thereby effectively improving the market competitiveness of products and improving customer satisfaction.

To achieve the above objectives, the present application proposes a product recommendation method, which comprises:

Obtaining a product recommendation request from a target customer group, wherein the product recommendation request includes a customer profile;

Based on the customer portrait, determine a product recommendation list corresponding to the target customer group;

Based on the product recommendation request and the product recommendation list, configure a product recommendation strategy;

Based on the product recommendation strategy, product recommendations are made to the target customer group.

In one embodiment, before obtaining the product recommendation request of the target customer group, the method further includes:

Obtain historical transaction records and recent visit information of target customers;

Inputting the historical transaction record into a customer portrait generation model to obtain a first portrait label output by the customer portrait generation model;

Based on the recent access information, generate a second portrait tag;

The first portrait tag and the second portrait tag are associated and combined to generate a customer portrait corresponding to the target customer.

In one embodiment, before inputting the historical transaction records and the recent access information into a customer profile generation model to obtain a customer profile output by the customer profile generation model, the method further includes:

Obtain historical transaction records of several customers;

Performing data screening on each of the historical transaction records;

The screened historical transaction records are input into the initial customer profile generation model for iterative training to obtain the customer profile generation model.

In one embodiment, determining a product recommendation list corresponding to the target customer group based on the customer portrait includes:

Based on the customer portrait, obtaining customer group information corresponding to the target customer group;

Based on the product recommendation tendency of the customer group information, a number of target products are obtained;

The target products are associated and combined to generate a product recommendation list corresponding to the target customer group.

In one embodiment, configuring a product recommendation strategy based on the product recommendation request and the product recommendation list includes:

Get the recommended layout corresponding to the target customer group;

Determine a product label corresponding to each target product in the product recommendation list;

Determine a product recommendation ranking based on each of the product tags and the customer portrait in the product recommendation request;

Based on the recommendation layout and the product recommendation ranking, a product recommendation strategy is configured.

In one embodiment, determining the product recommendation ranking based on each of the product tags and the customer portrait in the product recommendation request includes:

Determine the similarity between each of the product labels and the portrait label in the customer portrait;

Sorting the similarities in reverse order to obtain a similarity ranking;

The similarity ranking is used as the product recommendation ranking.

In one embodiment, after configuring the product recommendation strategy based on the product recommendation list, the method further includes:

Obtaining the product recommendation effect corresponding to the product recommendation strategy;

Based on the product recommendation effect, generate a strategy optimization report to adjust the product recommendation strategy;

Based on the adjusted product recommendation strategy, product recommendations are made to the target customer group.

In addition, to achieve the above purpose, the present application also proposes a product recommendation device, which includes:

An acquisition module, used to acquire a product recommendation request from a target customer group, wherein the product recommendation request includes a customer profile;

A determination module, used to determine a product recommendation list corresponding to the target customer group based on the customer portrait;

A configuration module, configured to configure a product recommendation strategy based on the product recommendation request and the product recommendation list;

A recommendation module is used to recommend products to the target customer group based on the product recommendation strategy.

In addition, to achieve the above-mentioned purpose, the present application also proposes a product recommendation device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the product recommendation method described above.

In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the product recommendation method described above are implemented.

In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the product recommendation method described above.

The present application provides a product recommendation method, apparatus, device and storage medium. The product recommendation method obtains a product recommendation request from a target customer group, wherein the product recommendation request includes a customer portrait, and then based on the customer portrait, determines a product recommendation list corresponding to the target customer group, thereby configuring a product recommendation strategy based on the product recommendation request and the product recommendation list, and then based on the product recommendation strategy, recommends products to the target customer group, thereby providing personalized product recommendations based on the actual needs and purchasing preferences of each customer, thereby improving the accuracy and success rate of product recommendations, thereby effectively improving the market competitiveness of the product and improving customer satisfaction.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

FIG1 is a schematic diagram of a process flow diagram provided in Example 1 of a product recommendation method of the present application;

FIG2 is a schematic diagram of a flow chart of a second embodiment of a product recommendation method of the present application;

FIG3 is a flow chart of a third embodiment of the product recommendation method of the present application;

FIG4 is a flowchart illustrating a strategy optimization report generation process in an embodiment of the present application;

FIG5 is a schematic diagram of the module structure of the product recommendation device according to an embodiment of the present application;

FIG6 is a schematic diagram of the device structure of the hardware operating environment involved in the product recommendation method in the embodiment of the present application.

The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

DETAILED DESCRIPTION

It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a big data service platform, a product recommendation system, etc. The following takes the product recommendation system as an example to illustrate this embodiment and the following embodiments.

Based on this, an embodiment of the present application provides a product recommendation method, referring to Figure 1, which is a flow chart of Example 1 of the product recommendation method of the present application.

In this embodiment, the product recommendation method includes steps S11 to S14:

Step S11, obtaining a product recommendation request from a target customer group, wherein the product recommendation request includes a customer portrait;

It should be noted that the target customer group refers to the customer group formed by associating and combining target customers according to different customer portrait tags, such as the customer group with "low risk preference", etc., which is not limited here. In addition, the product recommendation request refers to the request triggered when recommending products to the target customer group, including customer information (such as customer name, customer ID, etc.), customer portrait, and request time, etc., which is not limited here.

It should be further explained that the customer portrait is a tool used for marketing and customer management. It forms a comprehensive description and analysis of customers by collecting and analyzing various customer data, and is used to understand the customer's personal characteristics, needs, behavioral habits and preferences, so as to better serve customers and formulate corresponding marketing strategies. The customer portrait is usually generated by basic customer information (such as name, service usage time, customer background, etc.), historical purchase records, and portrait tags, etc., to achieve personalized customization and accurate recommendations, and improve customer satisfaction and loyalty. In addition, according to the business needs of the enterprise, the customer portrait can also be generated by information such as the customer's lifestyle, social relations, health status, etc., which is not limited here.

Specifically, product recommendation requests of the target customer group can be obtained, for example, by triggering the generation of a product recommendation request when a customer in the target customer group opens a business APP, or by triggering the generation of a product recommendation request when a customer in the target customer group refreshes a related section of the product recommendation. The system can also trigger the generation of a product recommendation request corresponding to the target customer group on its own, so as to configure the product recommendation strategy in advance and improve the efficiency of product recommendation. There is no restriction on the acquisition method here, and it can be set according to actual conditions.

Step S12, determining a product recommendation list corresponding to the target customer group based on the customer portrait;

It should be noted that the product recommendation list refers to a combined list of multiple target products that match the customer profile, including the product name, product label, and product details of the target product.

Specifically, based on the customer portrait, customer information corresponding to the target customer group is obtained, and then based on the product recommendation tendency of the customer information, a number of target products are obtained, so that the target products are associated and combined to generate a product recommendation list corresponding to the target customer group.

Step S13, configuring a product recommendation strategy based on the product recommendation request and the product recommendation list;

It should be noted that the product recommendation strategy refers to the recommendation strategy adopted when recommending products to the target customer group, including the product recommendation list, recommendation method, recommendation layout and customer group information.

Specifically, the recommendation layout corresponding to the target customer group is obtained, and then the product tags corresponding to each target product in the product recommendation list are determined, so as to determine the product recommendation ranking based on each product tag and the customer portrait in the product recommendation request, and then the product recommendation strategy is configured based on the recommendation layout and the product recommendation ranking.

Step S14: recommend products to the target customer group based on the product recommendation strategy.

Specifically, according to the product recommendation strategy, product recommendations are made to the target customer group. For example, if the product recommendation strategy is "push the product recommendation list of product A, product B and product C in the current product order on the 'home page', 'wealth zone section' and 'transaction center section'", then according to the product recommendation strategy, products are displayed in the push channels of the target customer group, wherein the push channels include APP, mini-programs and web pages, etc., which are not limited here and can be set according to actual conditions.

This embodiment obtains product recommendation requests from the target customer group, wherein the product recommendation request includes a customer portrait, and then based on the customer portrait, determines a product recommendation list corresponding to the target customer group, and then configures a product recommendation strategy based on the product recommendation request and the product recommendation list, and then based on the product recommendation strategy, recommends products to the target customer group, thereby providing personalized product recommendations based on the actual needs and purchasing preferences of each customer, thereby improving the accuracy and success rate of product recommendations, and then effectively improving the market competitiveness of products and enhancing customer satisfaction.

Based on this, an embodiment of the present application provides a product recommendation method, referring to Figure 2, which is a flow chart of Example 2 of the product recommendation method of the present application.

In a feasible implementation manner, before obtaining the product recommendation request of the target customer group, the method further includes:

Step S21, obtaining the target customer's historical transaction records and recent visit information;

It should be noted that the target customers refer to the customers in the target customer group. The historical transaction records refer to the historical business transaction records between the target customers and the merchants, including transaction business information (such as business type, etc.), transaction amount, transaction time, transaction frequency, transaction method (such as online payment, cash payment, etc.) and transactions, etc., which help merchants to establish accurate customer portraits based on customers' consumption behaviors and preferences.

It should be further explained that the recent visit information refers to the target customer's recent interactive behavior record with the product and business, for example, the date, time, pages viewed, dwell time, search history, click behavior, etc. when the customer visits the merchant's website or application.

Specifically, the data may be obtained by organizing and analyzing various business data of the target customer in a data warehouse or database, or by analyzing and obtaining log information generated by the target customer's operation in the system, without limitation here.

Step S22, inputting the historical transaction record into a customer portrait generation model to obtain a first portrait label output by the customer portrait generation model;

It should be noted that the customer portrait generation model is used to generate portrait tags corresponding to target customers. The first portrait tag refers to a customer portrait tag generated based on historical transaction records, such as "customers whose products expire within 15 days", "customers who have lost wealth management and deposits", etc., to characterize the current business status of the target customer and achieve accurate identification of customer tags.

Step S23, generating a second portrait tag based on the recent access information;

It should be noted that the second portrait label refers to a customer portrait label generated based on recent visit information, for example, "the number of visits to the wealth management area in the past month".

Specifically, the second profile tag can be generated by statistically analyzing the recent access information generated by the target customer within a preset time. For example, if the target customer has searched for "low-risk financial products" in the last three days, "low-risk financial products" can be used as the second profile tag of the target customer. The preset time can be the last three days, the last week, etc., and can be set according to actual conditions, and is not limited here.

Step S24: Associating and combining the first portrait tag and the second portrait tag to generate a customer portrait corresponding to the target customer.

It should be noted that the customer portrait changes with the target customer's historical transaction records and recent access information. The system can update the customer portrait to ensure the timeliness of the customer portrait. In addition, by updating the customer portrait when the customer is offline, business interruptions and the like can be prevented from occurring, which may affect normal business operations.

Specifically, the first portrait tag and the second portrait tag are associated and combined to generate a customer portrait corresponding to the target customer. In order to improve the accuracy and precision of customer portrait generation, the first portrait tag and the second portrait tag can be expanded through semantic analysis technology to obtain label words with similar semantics to the first portrait tag and the second portrait tag, thereby preventing the failure to recommend products suitable for customers due to the narrowing of the scope of customer portrait generation.

This embodiment obtains the historical transaction records and recent visit information of the target customer, and then inputs the historical transaction records into the customer portrait generation model to obtain the first portrait tag output by the customer portrait generation model, and then generates a second portrait tag based on the recent visit information, and then associates and combines the first portrait tag with the second portrait tag to generate a customer portrait corresponding to the target customer, thereby constructing a more complete, accurate and professional customer portrait by analyzing the customer's consumption habits, preferences and behavior patterns, and then providing more personalized and accurate product recommendations and services, and effectively improving the product sales success rate and customer loyalty.

In a feasible implementation manner, before inputting the historical transaction records and the recent access information into the customer portrait generation model to obtain the customer portrait output by the customer portrait generation model, the method further includes:

Step S31, obtaining historical transaction records of several customers;

Specifically, it is obtained through the business database of the merchant, which is not limited here and can be set according to actual conditions.

Step S32, screening the historical transaction records;

It should be noted that, in one embodiment, for banks, if they directly analyze the historical transaction records of customers, they will find that some customers will fill in transfer summaries such as "subscribe to financial products" and "subscribe to fund products" when transferring funds to other banks. Such customers have obvious financial management intentions in other banks and can be marked with customer portrait labels of "transferring to other banks to purchase wealth products". When identifying such customers, if only the keywords "subscription", "financial management" and "fund" are used to perform simple customer portrait label identification on the transfer summary, it is easy to generate a large number of invalid labels, such as "subscribe to buildings", "education funds", "housing maintenance funds", "financial management interest return", etc. Such invalid labels will interfere with the accuracy of product recommendations for customers such as "transferring to other banks to purchase wealth products" and affect the marketing of account managers. Therefore, it is necessary to filter such invalid labels to improve the accuracy and precision of customer portrait generation.

Specifically, the data of each of the historical transaction records can be screened by semantic analysis technology, or by setting preset screening rules. There is no limitation here and it can be set according to actual conditions.

Step S33, inputting the screened historical transaction records into the initial customer portrait generation model for iterative training to obtain the customer portrait generation model.

It should be noted that the customer portrait generation model is used to generate customer portrait labels, and can be a deep learning model, a machine learning model, a natural language processing (NLP) model, etc., and there is no limitation here.

It should be further explained that in order to further reduce the occurrence of invalid labels, in one embodiment, the customer portrait generation model can be a BERT (Bidirectional Encoder Representations from Transformers) model, which captures the deep bidirectional representation of language through pre-training on large-scale text data, thereby realizing feature extraction and intent analysis of text data, and then fine-tuning for different natural language processing tasks. It can filter out a large number of invalid labels, such as "subscribe to a building", "education fund", etc., which are invalid labels for customers such as "transferring money to other banks to purchase wealth products", greatly improving the accuracy of business scenario recognition.

Specifically, data preprocessing is performed on each filtered historical transaction record, such as data cleaning, normalization, chapter division, etc., and then each filtered historical transaction record and the sample label corresponding to the historical transaction record are input into the initial customer portrait generation model to obtain the predicted value output by the initial chapter customer portrait generation model, and then based on the predicted value and the sample label, the model loss value is calculated using the loss function. In this embodiment, the loss function can be set according to actual needs and is not specifically limited here. After the model loss value is calculated, the training process ends, and the error back propagation algorithm is used to update the model parameters in the initial customer portrait generation model, and then the next training is performed. During the training process, it is determined whether the updated initial customer portrait generation model meets the preset training end condition. If it meets, the updated initial customer portrait generation model is used as the customer portrait generation model. If it does not meet, the model training continues, wherein the preset training end condition includes loss convergence and reaching the maximum number of iterations. The setting can be made according to the actual situation and is not limited here, thereby improving the efficiency of customer unstructured data recognition.

This embodiment obtains historical transaction records of several customers and then performs data screening on each of the historical transaction records, thereby inputting the screened historical transaction records into an initial customer portrait generation model for iterative training to obtain the customer portrait generation model, thereby forming a more accurate customer portrait, increasing the success rate of product recommendations, and reducing the cost of manual identification by business personnel.

Based on this, an embodiment of the present application provides a product recommendation method, referring to Figure 3, which is a flow chart of Example 3 of the product recommendation method of the present application.

In a feasible implementation manner, determining a product recommendation list corresponding to the target customer group based on the customer portrait includes:

Step S41, obtaining customer group information corresponding to the target customer group based on the customer portrait;

It should be noted that the customer group information refers to the customer group-related information of the target customer group, including customer group labels, customer information, and product recommendation tendencies.

Specifically, based on the customer portrait, the customer information corresponding to the target customer group is obtained. For example, the portrait label in the customer portrait is used as the customer label of the target customer group, and the product recommendation tendency is determined based on the customer label. For example, when the customer label is "the proportion of time spent browsing low-risk products within three days is 60%", the product recommendation tendency is "low-risk products". There is no restriction here and it can be set according to actual conditions.

In addition, because the customer portrait of each customer will be refreshed, the customers included in the target customer group will also be refreshed, and the customer group to which each customer belongs will also be different, thereby achieving a recommendation method that is tailored to each individual.

Step S42, obtaining a plurality of target products based on the product recommendation tendency of the customer group information;

It should be noted that the product recommendation tendency refers to the products preferred by the target customer group, such as products with a risk level less than R2, products with a seven-day annualized return greater than 2%, etc. In addition, the target product refers to the product that meets the needs of the target customer group, including product name, product introduction and product label.

Specifically, based on the product recommendation tendency of the customer information, several target products are obtained. For example, the product recommendation tendency is matched with a preset product database to obtain several target products. As another example, the product recommendation tendency is pushed to professionals, such as product managers and account managers of a bank, so that the professionals can screen products and obtain several target products. In this way, by combining the analysis and experience of professionals on customers, products, and markets, the recommendation logic of both online and offline recommendation channels can be unified and professionalized, thereby improving customer trust.

Step S43: associate and combine the target products to generate a product recommendation list corresponding to the target customer group.

Specifically, the product information of each of the target products is associated and combined to generate a product recommendation list corresponding to the target customer group.

This embodiment obtains customer information corresponding to the target customer group based on the customer portrait, and then obtains several target products based on the product recommendation tendency of the customer information, so as to associate and combine the target products to generate a product recommendation list corresponding to the target customer group, thereby improving the accuracy of personalized recommendations, increasing product sales conversion rate and market competitiveness, and thus improving customer satisfaction and loyalty.

In a feasible implementation manner, configuring a product recommendation strategy based on the product recommendation request and the product recommendation list includes:

Step S51, obtaining the recommended layout corresponding to the target customer group;

It should be noted that the recommended layout refers to the content section or page recommended to target customers based on their usage interests and needs, such as the financial management area in a bank APP, etc., and there is no restriction here.

Specifically, the recommended layout corresponding to the target customer group can be obtained through the clicks and browsing behaviors of each target customer in the target customer group. For example, when the time a customer spends browsing product information on the APP homepage accounts for more than 40% of the total APP usage time, the APP homepage will be used as the recommended layout. There is no restriction on the acquisition method, which can be set according to actual conditions.

Step S52, determining the product label corresponding to each target product in the product recommendation list;

It should be noted that the product label refers to an information label used to identify and classify products, such as high-risk and high-return products, so as to quickly identify the differences between different products and their corresponding advantages and disadvantages.

Specifically, the product label corresponding to each target product in the product recommendation list may be determined by querying the data information of the target product, which is not limited here.

Step S53, determining a product recommendation ranking based on each of the product tags and the customer portrait in the product recommendation request;

It should be noted that the product recommendation ranking is used to sort the product list recommended to the user to highlight the products that are most likely to meet customer needs or are most relevant, thereby improving user experience and recommendation effects.

Specifically, the similarity between each product label and the portrait label in the customer portrait is determined, and then each similarity is sorted in reverse order to obtain a similarity ranking, and the similarity ranking is used as the product recommendation ranking.

Step S54, configuring a product recommendation strategy based on the recommendation layout and the product recommendation ranking.

Specifically, the product recommendation strategy may be: according to the product recommendation ranking, the product information of each target product is sorted, a product recommendation page is generated, and then the product recommendation page is displayed on the recommendation page, wherein, the customer can be attracted to view by adding display styles, for example, target products with high recommendation thresholds are displayed independently, etc., so that the target customers can view and purchase them. There is no restriction on the configuration method here, and it can be set according to actual conditions.

This embodiment obtains the recommendation layout corresponding to the target customer group, and then determines the product tags corresponding to each target product in the product recommendation list, and then determines the product recommendation ranking based on each product tag and the customer portrait in the product recommendation request, and then configures the product recommendation strategy based on the recommendation layout and the product recommendation ranking, so as to effectively identify and locate customer needs, provide more personalized product recommendations, improve the accuracy and success rate of product recommendations, and effectively improve the market competitiveness of products and enhance customer satisfaction.

In a feasible implementation manner, determining the product recommendation ranking based on each of the product tags and the customer portrait in the product recommendation request includes:

Step S61, determining the similarity between each of the product labels and the portrait label in the customer portrait;

It should be noted that the similarity is used to characterize the similarity between the product label and the image label.

Specifically, the similarity can be calculated by different algorithms and indicators. For example, each product label and the portrait label in the customer portrait are vectorized, and then the similarity between each vectorized product label and the portrait label in the customer portrait is calculated by algorithms such as cosine similarity and Euclidean distance.

In addition, the semantic analysis model can also be used to understand the labels of each product label and the portrait label in the customer portrait, so as to accurately identify the meaning of the label, reduce ambiguity, ensure that the interpretation of the label is accurate, and then analyze the semantic content and emotional tendency of the label, and evaluate the similarity between different labels. There is no restriction on the method of determining the similarity, which can be set according to actual conditions.

Step S62, sorting the similarities in reverse order to obtain a similarity ranking;

It should be noted that the similarity ranking refers to the ranking of the similarities between product labels and image labels from high to low.

Specifically, the similarities are sorted in reverse order, that is, from high similarity to low similarity, so as to obtain a similarity sorting.

Step S63: using the similarity ranking as the product recommendation ranking.

This embodiment determines the similarity between each product tag and the portrait tag in the customer portrait, and then sorts the similarities in reverse order to obtain a similarity ranking, and then uses the similarity ranking as the product recommendation ranking, thereby ensuring that the recommendation system shows users the products they are most likely to be interested in, thereby improving user experience and purchase likelihood and providing personalized product recommendations.

In a feasible implementation manner, after configuring the product recommendation strategy based on the product recommendation list, the method further includes:

Step S71, obtaining the product recommendation effect corresponding to the product recommendation strategy;

It should be noted that the product recommendation effect refers to the recommendation effect achieved within a certain period of time after recommending a product or service to the target customer, such as the number of clicks, number of purchases, purchase amount, recommendation conversion rate, etc. of the customer after recommending the product.

Specifically, event tracking (Event Tracking refers to the related technology and implementation process for capturing, processing and sending specific customer behaviors or events) is performed on the transaction operations of the target customers to analyze and identify the number of clicks, number of purchases, purchase amount, etc. of the target products corresponding to the product recommendation strategy, so as to obtain the product recommendation effect corresponding to the product recommendation strategy.

In addition, the product recommendation effect corresponding to the product recommendation strategy can be obtained by performing statistical analysis on customer operation records in the system. There is no restriction on the acquisition method, which can be set according to actual conditions.

Step S72, generating a strategy optimization report based on the product recommendation effect to adjust the product recommendation strategy;

It should be noted that the strategy optimization report refers to a detailed report generated when optimizing the product recommendation strategy, including recommendation strategy information, strategy performance indicators (such as accuracy, recall rate, F1 score, etc.), customer behavior analysis (such as click-through rate, purchase rate, satisfaction, etc.), recommendation result analysis (such as selection, sorting and combination of recommended products, etc.), and problems encountered in the recommendation process and optimization suggestions, etc., please refer to Figure 4, which is an example diagram of the process of generating a strategy optimization report in an embodiment of the present application.

Specifically, a strategy optimization report is generated based on the product recommendation effect, wherein the product recommendation effect can be analyzed by an analysis tool to generate a strategy optimization report, or the product recommendation effect and the product recommendation strategy are input into a recommendation strategy optimization model for analysis to obtain a strategy optimization report output by the recommendation strategy optimization model, wherein the recommendation strategy optimization model is iteratively generated based on historical product recommendation strategies and the product recommendation effects corresponding to the historical product recommendation strategies. Further, the product recommendation strategy is adjusted based on the strategy optimization report.

In addition, if the effect rating of the product recommendation effect is good, the product recommendation effect is associated with the target product with good effect for recommendation, so as to increase the credibility of the product effect and improve the product purchase rate. In addition, if the effect rating of the product recommendation effect is poor, the product recommendation effect is associated with the target product with poor effect and pushed to professionals for professional product analysis, so as to reduce the recommendation of invalid products and reduce the waste of resources and costs.

Step S73: recommend products to the target customer group based on the adjusted product recommendation strategy.

Specifically, according to the adjusted product recommendation strategy, products are recommended to the target customer group, and the effect evaluation and strategy optimization of the product recommendation strategy are continuously performed, thereby forming a recommendation service closed loop and continuously improving the recommendation effect of the customer recommendation strategy.

This embodiment obtains the product recommendation effect corresponding to the product recommendation strategy, and then generates a strategy optimization report based on the product recommendation effect to adjust the product recommendation strategy, so as to make product recommendations to the target customer group based on the adjusted product recommendation strategy, thereby realizing data-driven decision-making to better adapt to market changes and changes in customer behavior, enhance market adaptability, and optimize product management, thereby increasing the market competitiveness of products and customer satisfaction with recommended products.

It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean 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 embodiment of the present invention.

The present application also provides a product recommendation device, please refer to FIG. 5 , the product recommendation device includes:

An acquisition module 51 is used to acquire a product recommendation request from a target customer group, wherein the product recommendation request includes a customer profile;

A determination module 52, configured to determine a product recommendation list corresponding to the target customer group based on the customer portrait;

A configuration module 53, configured to configure a product recommendation strategy based on the product recommendation request and the product recommendation list;

The recommendation module 54 is used to make product recommendations to the target customer group based on the product recommendation strategy.

Product recommendation devices are also used to:

Obtain historical transaction records and recent visit information of target customers;

Inputting the historical transaction record into a customer portrait generation model to obtain a first portrait label output by the customer portrait generation model;

Based on the recent access information, generate a second portrait tag;

The first portrait tag and the second portrait tag are associated and combined to generate a customer portrait corresponding to the target customer.

Product recommendation devices are also used to:

Obtain historical transaction records of several customers;

Performing data screening on each of the historical transaction records;

The screened historical transaction records are input into the initial customer profile generation model for iterative training to obtain the customer profile generation model.

Product Recommendation Devices are also used to:

Based on the customer portrait, obtaining customer group information corresponding to the target customer group;

Based on the product recommendation tendency of the customer group information, a number of target products are obtained;

The target products are associated and combined to generate a product recommendation list corresponding to the target customer group.

Product recommendation devices are also used to:

Get the recommended layout corresponding to the target customer group;

Determine a product label corresponding to each target product in the product recommendation list;

Determine a product recommendation ranking based on each of the product tags and the customer portrait in the product recommendation request;

Based on the recommendation layout and the product recommendation ranking, a product recommendation strategy is configured.

Product recommendation devices are also used to:

Determine the similarity between each of the product labels and the portrait label in the customer portrait;

Sorting the similarities in reverse order to obtain a similarity ranking;

The similarity ranking is used as the product recommendation ranking.

Product recommendation devices are also used to:

Obtaining the product recommendation effect corresponding to the product recommendation strategy;

Based on the product recommendation effect, generate a strategy optimization report to adjust the product recommendation strategy;

Based on the adjusted product recommendation strategy, product recommendations are made to the target customer group.

The product recommendation device provided by the present application adopts the product recommendation method in the above embodiment, which can solve the technical problems in the background technology. Compared with the prior art, the beneficial effects of the product recommendation device provided by the present application are the same as the beneficial effects of the product recommendation method provided by the above embodiment, and the other technical features in the product recommendation device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

The present application provides a product recommendation device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the product recommendation method in the above-mentioned embodiment one.

Reference is made to Figure 6 below, which shows a schematic diagram of the structure of a product recommendation device suitable for implementing an embodiment of the present application. The product recommendation device in the embodiment of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The product recommendation device shown in Figure 6 is only an example and should not bring any limitations to the functions and scope of use of the embodiments of the present application.

As shown in FIG6 , the product recommendation device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the product recommendation device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input/output (I/O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I/O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the product recommendation device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a product recommendation device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.

In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

The product recommendation device provided by the present application adopts the product recommendation method in the above embodiment, which can solve the technical problems in the background technology. Compared with the prior art, the beneficial effects of the product recommendation device provided by the present application are the same as the beneficial effects of the product recommendation method provided by the above embodiment, and the other technical features in the product recommendation device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the product recommendation method in the above-mentioned embodiment.

The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

The computer-readable storage medium may be included in the product recommendation device; or may exist independently without being assembled into the product recommendation device.

The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the product recommendation device, the product recommendation device:

Obtaining a product recommendation request from a target customer group, wherein the product recommendation request includes a customer profile;

Based on the customer portrait, determine a product recommendation list corresponding to the target customer group;

Based on the product recommendation request and the product recommendation list, configure a product recommendation strategy;

Based on the product recommendation strategy, product recommendations are made to the target customer group.

Computer program code for performing the operations of the present application may be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the client computer, partially on the client computer, as a separate software package, partially on the client computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the client computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and/or flow chart, and the combination of the square boxes in the block diagram and/or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned product recommendation method, and can solve the technical problems in the background technology. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the product recommendation method provided in the above-mentioned embodiment, and will not be repeated here.

An embodiment of the present application provides a computer program product, including a computer program, which implements the steps of the product recommendation method as described above when executed by a processor.

The computer program product provided in this application can solve the technical problems in the background technology. Compared with the prior art, the beneficial effects of the computer program product provided in the embodiment of this application are the same as the beneficial effects of the product recommendation method provided in the above embodiment, which will not be repeated here.

The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct/indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims (10)

1. A method of product recommendation, comprising:

obtaining a product recommendation request of a target guest group, wherein the product recommendation request comprises a customer portrait;

Determining a product recommendation list corresponding to the target guest group based on the guest representation;

Configuring a product recommendation strategy based on the product recommendation request and the product recommendation list;

and recommending the products to the target guest group based on the product recommendation strategy.

2. The product recommendation method as claimed in claim 1, wherein prior to said obtaining the product recommendation request for the target group of guests, further comprising:

Acquiring historical transaction records and recent access information of each client in a target client group;

Inputting the historical transaction record into a customer portrait generation model to obtain a first portrait tag output by the customer portrait generation model;

Generating a second portrait tag based on the recent access information;

and associating and combining the first portrait tag and the second portrait tag to generate a customer portrait corresponding to each customer.

3. The product recommendation method of claim 2, wherein before inputting the historical transaction record and the recent access information into a customer representation generation model to obtain a customer representation output by the customer representation generation model, further comprising:

Acquiring historical transaction records of a plurality of clients;

data screening is carried out on each historical transaction record;

And inputting each history transaction record after screening into an initial customer portrait generation model for iterative training to obtain the customer portrait generation model.

4. The product recommendation method as claimed in claim 1, wherein said determining a product recommendation list corresponding to said target group based on said customer representation comprises:

Based on the customer portrait, obtaining guest group information corresponding to the target guest group;

obtaining a plurality of target products based on the product recommendation tendency of the guest group information;

and carrying out association combination on each target product to generate a product recommendation list corresponding to the target customer group.

5. The product recommendation method of claim 1, wherein said configuring a product recommendation policy based on said product recommendation request and said product recommendation list comprises:

Acquiring a recommended layout corresponding to a target guest group;

Determining product labels corresponding to all target products in the product recommendation list;

Determining a product recommendation order based on each of the product tags and the customer representation in the product recommendation request;

and configuring a product recommendation strategy based on the recommendation layout and the product recommendation sequence.

6. The product recommendation method of claim 5, wherein determining a product recommendation order based on each of the product tags and the customer representation in the product recommendation request comprises:

determining the similarity of each product label and the portrait label in the customer portrait;

Performing reverse order sorting on the similarity to obtain similarity sorting;

and taking the similarity ranking as the product recommendation ranking.

7. The product recommendation method of claim 1, wherein after configuring a product recommendation policy based on the product recommendation list, further comprising:

acquiring a product recommendation effect corresponding to the product recommendation strategy;

Generating a strategy optimization report based on the product recommendation effect so as to adjust the product recommendation strategy;

and recommending the products to the target guest group based on the adjusted product recommendation strategy.

8. A product recommendation device, comprising:

The system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring a product recommendation request of a target guest group, and the product recommendation request comprises a customer portrait;

the determining module is used for determining a product recommendation list corresponding to the target guest group based on the customer portrait;

The configuration module is used for configuring a product recommendation strategy based on the product recommendation request and the product recommendation list;

And the recommending module is used for recommending the products to the target guest group based on the product recommending strategy.

9. A product recommendation device, characterized in that the product recommendation device comprises: memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the product recommendation method according to any of claims 1 to 7.

10. A storage medium, characterized in that the storage medium is a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the steps of the product recommendation method according to any one of claims 1 to 7.