CN113935816A - Product Recommendation Methods, Apparatus, Electronic Equipment, Media and Program Products - Google Patents

Abstract

The disclosure provides a product recommendation method, and relates to the field of artificial intelligence or the field of finance. The product recommendation method comprises the following steps: acquiring a loyalty category corresponding to a user and a sales category of each first product in the N first products; obtaining corresponding evaluation indexes based on the sales amount category of each first product by using preset evaluation conditions; and determining M second products from the N first products based on the loyalty category to recommend to the user, wherein the evaluation index of each second product is greater than or equal to a preset threshold value. The present disclosure also provides a product recommendation apparatus, device, storage medium and program product.

Description

Product recommendation method, device, electronic equipment, medium and program product

Technical Field

The present disclosure relates to the field of artificial intelligence or the field of finance, and the like, and more particularly, to a product recommendation method, apparatus, electronic device, medium, and program product.

Background

With the development of society and the gradual expansion of economic globalization, product competition is more and more intense. Therefore, recommending high-quality products to users to achieve the purposes of attracting new users and retaining old users is one of means for improving competitiveness. The product recommendation can be to recommend products when the users carry out face-to-face marketing, or can be to make a marketing strategy in advance and then execute a product recommendation scheme on line or off line according to the marketing strategy so as to promote the users to know and use the recommended products.

In the related art, generally, business personnel recommend products to different users according to own experience. For example, a business person may recommend a corresponding product according to an impression of talking about, wearing, etc. of a new user, or may recommend a corresponding product according to a usage habit of an old user to promote a purchase intention of the user.

In the course of implementing the disclosed concept, the inventors found that there are at least the following problems in the prior art:

depending on the subjective experience of the service personnel, products cannot be objectively recommended to different users in a targeted manner. For example, different business personnel have different experiences and different recommended schemes, and business personnel cannot be familiar with numerous products one by one and cannot accurately grasp the requirements of different users.

Disclosure of Invention

In view of the above, the present disclosure provides a product recommendation method, apparatus, electronic device, medium, and program product that can perform targeted marketing to a user.

One aspect of the disclosed embodiments provides a product recommendation method, including: acquiring a loyalty category corresponding to a user and a sales category of each first product in the N first products; obtaining corresponding evaluation indexes based on the sales amount category of each first product by using preset evaluation conditions; and determining M second products from the N first products based on the loyalty category to recommend to the user, wherein the evaluation index of each second product is greater than or equal to a preset threshold, and N and M are integers greater than or equal to 1 respectively.

According to an embodiment of the present disclosure, said determining M second products from the N first products to recommend to the user based on the loyalty category comprises: under the condition that the user is authorized, acquiring historical data of products purchased by the user; determining the M second products based on the loyalty category and the historical data if the historical data is obtained.

According to an embodiment of the present disclosure, said determining the M second products based on the loyalty category and the historical data comprises: determining S third products from the historical data, wherein the N first products comprise the S third products, the evaluation index of each third product is greater than or equal to the preset threshold, and S is an integer greater than or equal to 1; determining the M second products based on the loyalty category, including: taking the S third products as the S second products, wherein S is less than or equal to M; or based on the product types of the S third products, determining M first products of the same type from the N first products as the M second products.

According to an embodiment of the present disclosure, the obtaining the loyalty category corresponding to the user includes: classifying the user through a user classification model to obtain the loyalty classification, wherein the method specifically comprises the following steps: acquiring first characteristic data of a user, wherein the first characteristic data comprises interaction data of the user and organizations to which the N first products belong; inputting the first feature data into the user classification model to obtain the loyalty classification.

According to an embodiment of the present disclosure, the interaction data includes at least one of: identity information of the user registered at the organization, rating information of the user at the organization, asset information of the user, historical data of the user purchasing products at the organization.

According to an embodiment of the present disclosure, the obtaining of the sales category of each first product includes: classifying each first product through a product classification model to obtain the sales amount category, wherein the sales amount category specifically includes: acquiring second characteristic data of each first product, wherein the second characteristic data comprises sales data of each first product; inputting the second characteristic data of each first product into the product classification model to predict the sales amount category corresponding to each first product.

According to an embodiment of the present disclosure, the sales data comprises at least one of: and the sales channel, the promotion times, the expected earning rate, the historical earning and the purchase mode of each first product.

According to an embodiment of the present disclosure, the obtaining, by using a preset evaluation condition, a corresponding evaluation index based on the sales amount category of each first product includes: acquiring feedback data of each first product, wherein the feedback data comprises user evaluation data of each first product in a sale process; and assigning a first weight to the feedback data and a second weight to the sales category to obtain the evaluation index, wherein the evaluation condition includes the first weight and the second weight.

Another aspect of the disclosed embodiments provides a product recommendation device, including: the system comprises a category acquisition module, an evaluation index module and a product recommendation module. The category obtaining module is used for obtaining a loyalty category corresponding to the user and a sales category of each first product in the N first products. The evaluation index module is used for obtaining corresponding evaluation indexes based on the sales amount category of each first product by using preset evaluation conditions. The product recommending module is used for determining M second products from the N first products based on the loyalty degree category so as to recommend the M second products to the user, wherein the evaluation index of each second product is larger than or equal to a preset threshold, and N and M are integers larger than or equal to 1 respectively.

Another aspect of the present disclosure provides an electronic device including: one or more processors; memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method as described above.

Another aspect of the present disclosure also provides a computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method as described above.

Another aspect of the disclosure also provides a computer program product comprising a computer program which, when executed by a processor, implements the method as described above.

One or more of the embodiments described above have the following advantages or benefits: the method can at least partially solve the problem that products cannot be objectively and pertinently recommended to different users by means of subjective experiences of service personnel, obtains the sales category of a first product by determining corresponding loyalty categories for different users, and then obtains corresponding evaluation indexes through the sales category of the first product by using preset evaluation conditions, so that second products with the evaluation indexes larger than or equal to a preset threshold value can be pertinently recommended to the users based on the loyalty categories of the users.

Drawings

The foregoing and other objects, features and advantages of the disclosure will be apparent from the following description of embodiments of the disclosure, which proceeds with reference to the accompanying drawings, in which:

FIG. 1 schematically illustrates an application scenario diagram suitable for implementing a product recommendation method according to an embodiment of the present disclosure;

FIG. 2 schematically illustrates a flow diagram of a method of product recommendation in accordance with an embodiment of the present disclosure;

FIG. 3 schematically illustrates a flow chart for determining M second products according to an embodiment of the present disclosure;

FIG. 4 schematically illustrates a flow diagram for determining M second products according to another embodiment of the present disclosure;

FIG. 5 schematically illustrates a flow diagram of a method of product recommendation, according to another embodiment of the present disclosure;

FIG. 6 schematically illustrates a flow diagram for earning loyalty categories, according to an embodiment of the present disclosure;

FIG. 7 schematically illustrates a flow chart for obtaining sales categories according to an embodiment of the present disclosure;

FIG. 8 schematically shows a flow chart for obtaining an evaluation index according to an embodiment of the present disclosure;

FIG. 9 is a block diagram schematically illustrating the structure of a product recommendation device according to an embodiment of the present disclosure; and

FIG. 10 schematically illustrates a block diagram of an electronic device suitable for implementing a method of product recommendation in accordance with an embodiment of the present disclosure.

Detailed Description

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is illustrative only and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.

In the related art, for example, in scenes of selling insurance products, vehicle products, e-commerce products, or financial products, intense competition is often faced, and product recommendation needs to be performed on marketing activities of users. Taking financial products of financial institutions as an example, when recommending financial products, business personnel may recommend several financial products to users subjectively according to the requirements learned in the process of communicating with the users. However, the risk bearing capability of the user may not be well known, or thousands of financial products may not be well known, so that the financial products recommended to the user may not be targeted, in other words, the requirements of the user may not be met.

Embodiments of the present disclosure provide a product recommendation method, apparatus, electronic device, medium, and program product. The product recommendation method comprises the following steps: and acquiring the loyalty category corresponding to the user and the sales category of each first product in the N first products. And obtaining a corresponding evaluation index based on the sales amount category of each first product by using a preset evaluation condition. And determining M second products from the N first products based on the loyalty category to recommend the M second products to the user, wherein the evaluation index of each second product is greater than or equal to a preset threshold, and N and M are integers greater than or equal to 1 respectively.

According to the embodiment of the disclosure, different loyalty categories corresponding to different users are obtained, the user characteristics of each loyalty category are obtained in a data mining mode, in addition, the N first products are evaluated respectively to obtain evaluation indexes, product recommendation can be performed on different users in a targeted mode, and the quality degree of recommended products received by the users can be improved.

It should be noted that, in some embodiments of the present disclosure, the product is, for example, anything that is provided to the market as a commodity, used and consumed by the user, and can meet the user's needs, such as a physical commodity or a professional service. In addition, the product recommendation method, the product recommendation device, the electronic device, the medium and the program product provided by the embodiment of the disclosure can be used in aspects of product recommendation, marketing scheme making and the like of big data and artificial intelligence technologies, and can also be used in various fields except big data counting and artificial intelligence technologies, such as financial fields and the like. The application fields of the product recommendation method, the product recommendation device, the electronic device, the medium and the program product provided by the embodiment of the disclosure are not limited.

In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision or application of the characteristic data of the related user all conform to the regulations of related laws and regulations, and the collection, storage, use, processing, transmission, provision or application of the characteristic data of the related user is executed under the condition of obtaining the authorization of the user, and necessary security measures are taken without violating the good custom of the public order.

Fig. 1 schematically illustrates an application scenario diagram suitable for implementing a product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 1, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

Clients or business persons, etc. may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages, etc. For example, the service person may send an instruction to the server 105 to acquire M second products through the terminal devices 101, 102, and 103, and the server 105 may implement the product recommendation method according to the embodiment of the present disclosure and return the M second products to the service person. For another example, the client accesses the server 105 through the terminal devices 101, 102, and 103, and the server 105 may implement the product recommendation method according to the embodiment of the present disclosure to display M second products to the client.

The terminal devices 101, 102, 103 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only). The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 105 may be a server providing various services, such as a background management server (for example only) providing support for websites browsed by users using the terminal devices 101, 102, 103. The background management server may analyze and perform other processing on the received data such as the user request, and feed back a processing result (e.g., a webpage, information, or data obtained or generated according to the user request) to the terminal device.

It should be noted that the product recommendation method provided by the embodiment of the present disclosure may be generally executed by the server 105. Accordingly, the product recommendation device provided by the embodiments of the present disclosure may be generally disposed in the server 105. The product recommendation method provided by the embodiment of the present disclosure may also be executed by a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and/or the server 105. Accordingly, the product recommendation device provided by the embodiment of the present disclosure may also be disposed in a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and/or the server 105.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

The product recommendation method according to the embodiment of the present disclosure will be described in detail below with reference to fig. 2 to 8 based on the scenario described in fig. 1.

FIG. 2 schematically shows a flow diagram of a product recommendation method according to an embodiment of the disclosure.

As shown in fig. 2, the product recommendation method of this embodiment includes operations S210 to S230.

In operation S210, a loyalty category corresponding to the user and a sales category of each of the N first products are obtained.

According to an embodiment of the present disclosure, the user herein refers to a customer of an organization to which the N first products belong. For example, the loyalty category may include a high loyalty category, a medium loyalty category, a low loyalty category, or a zero loyalty category. For example, users with high loyalty groups have a great degree of trust in the organization and have a high level of interest in repurchasing products once purchased, and are willing to actively learn about and try other products. Users of the medium loyalty class have a general level of trust with the organization and a moderate level of intent to re-purchase products once purchased and to try when passively receiving recommended products. Users with low loyalty categories, for example, have not created a level of trust with the organization, are familiar with the organization, and have purchased a small number of products and are willing to learn the same type of product as the products that have been purchased. For example, a user with zero loyalty may be a good product with a high evaluation index to a customer who knows the products of the organization but does not purchase the products.

It should be noted that the names and meanings of the high loyalty class, the medium loyalty class, the low loyalty class, or the zero loyalty class are only examples, and the users of each class can be flexibly defined according to practical applications, for example, the users can be flexibly classified according to questionnaires, product purchasing limits, product purchasing types, basic information (such as academic calendars, property and other information) of the users, and the like.

In operation S220, a corresponding evaluation index is obtained based on the sales amount category of each first product using a preset evaluation condition.

For example, the sales categories may include a high sales category (e.g., sales over 1000 ten thousand yuan), a medium sales category (e.g., sales between 500 ten thousand yuan and 1000 ten thousand yuan), and a low sales category (e.g., sales between 0 yuan and 500 ten thousand yuan), and the popularity of the current market for each first product may be reflected by different sales, so as to be used as a factor for calculating the evaluation index, and the high-quality product may be determined by the evaluation index.

According to an embodiment of the present disclosure, the evaluation condition may include four sub-evaluation conditions respectively corresponding to the high loyalty class, the medium loyalty class, the low loyalty class, or the zero loyalty class, for example, adjusting weights of different types of products based on the loyalty, and applying different calculation formulas. And screening out products matched with the loyalty categories by using the evaluation indexes by applying the sub-evaluation conditions of different categories. In other embodiments of the present disclosure, the evaluation condition includes a condition by which the evaluation index obtained is applicable to different loyalty categories.

In operation S230, M second products are determined from the N first products based on the loyalty category to recommend to the user, wherein an evaluation index of each second product is greater than or equal to a preset threshold, and N and M are integers greater than or equal to 1, respectively.

According to the embodiment of the disclosure, the effect of enabling the evaluation index of each second product to be larger than or equal to the preset threshold value is that a good-quality product can be screened out from the N first products, and the benefit of a user after purchasing the product is improved. For example, a good quality periodic financing product may be a product with a more stable return and a higher return (by way of example only). The preset threshold may be the top 20% (for example only) of the ranking based on the evaluation index.

In some embodiments, the ratings of each first product may be obtained based on sub-ratings corresponding to different loyalty categories, e.g., a first product may have ratings obtained using four sub-ratings. And determining corresponding sub-evaluation conditions based on the loyalty degree category of the user, and then obtaining M second products with the evaluation indexes larger than or equal to a preset threshold value.

In other embodiments, the evaluation criteria for each first product may be obtained by an evaluation condition, applicable to different loyalty categories. For example, a user with high loyalty may recommend a product with a high evaluation index, and may set a purchase condition as a special product to provide feedback to the user.

According to the embodiment of the disclosure, different loyalty categories corresponding to different users are obtained, the user characteristics of each loyalty category are obtained in a data mining mode, in addition, the N first products are evaluated respectively to obtain evaluation indexes, product recommendation can be performed on different users in a targeted mode, and the quality degree of recommended products received by the users can be improved. For an organization, the targeted product recommendation can improve the sales volume of the product, thereby bringing greater profit to the organization.

Fig. 3 schematically shows a flowchart of determining M second products in operation S230 according to an embodiment of the present disclosure.

As shown in fig. 3, determining M second products from the N first products based on the loyalty program in operation S230 includes operations S310 to S320.

In operation S310, in case of the user authorization, history data of the product purchased by the user is acquired.

In operation S320, in the case where the history data is acquired, M second products are determined based on the loyalty category and the history data.

According to the embodiment of the disclosure, if the user corresponds to the zero loyalty class, the historical data of the purchased product may not be obtained, and in this case, the second product with one or more evaluation indexes greater than or equal to the preset threshold value may be recommended to the user.

According to the embodiment of the disclosure, if the user corresponds to any one of the high loyalty class, the medium loyalty class and the low loyalty class, the requirement of the user can be better known from the historical data, so that the recommendation can be made in a targeted manner. For the different loyalty categories, the historical data may generate different reference functions, which may be described in detail with reference to fig. 4.

Fig. 4 schematically shows a flowchart of determining M second products in operation S320 according to another embodiment of the present disclosure.

As shown in fig. 4, in the case that the history data is acquired in operation S320, determining M second products based on the loyalty category and the history data includes operations S410 to S420, where operation S420 may include operation S421 or operation S422.

In operation S410, S third products are determined from the historical data, where the N first products include S third products, an evaluation index of each third product is greater than or equal to a preset threshold, and S is an integer greater than or equal to 1.

According to the embodiment of the disclosure, taking the evaluation index of each first product suitable for different loyalty categories as an example, the N first products may include all the products of the affiliated organization, and the third product purchased by the user is also within the range of the N first products. In addition, the evaluation index for each first product may be dynamically obtained within a predetermined time, for example, once per month. Therefore, on the basis of obtaining historical data, the mechanism screens the purchased products recommended by the user, improves the quality degree of the recommended products, and guarantees the benefit of the user.

In operation S420, M second products are determined based on the loyalty category.

In operation S421, S third products are regarded as S second products, where S is less than or equal to M.

According to the embodiment of the disclosure, if the user corresponds to the high loyalty class or the medium loyalty class, the users of the two classes have relatively high acceptance of the purchased product, and the third product can be directly recommended. Due to the different degrees of acceptance, the proportion of the third product may be adjusted adaptively, for example, a higher proportion (e.g., 60%, by way of example only) of the second products are recommended by users with a high loyalty class, and a lower proportion (e.g., 40%, by way of example only) of the second products are recommended by users with a medium loyalty class.

In some embodiments of the present disclosure, if the user corresponds to a low loyalty class, a third product having a lower proportion (e.g., 20%, by way of example only) may also be identified.

In operation S422, M first products of the same type are determined as M second products from the N first products based on the product types of the S third products.

Taking the financial product as an example, the first product may include regular products, bond products, fund products, and the like, and the regular products of the same type are regular products of different time, such as regular 3 months, regular 1 year, or regular 3 years (for example only).

According to embodiments of the present disclosure, for example, low loyalty users may purchase fewer products and may have difficulty creating an effective product recommendation, but would like to know the same type of product. Therefore, M first products of the same type may be determined to be recommended based on the product types of the S third products, where the M first products of the same type may include one or more of the S third products, and may specifically be determined according to the level of the evaluation index.

In other embodiments of the present disclosure, if the user corresponds to the high loyalty class or the medium loyalty class, based on the S third products as the S second products, operation S422 may also be performed to determine M-S second products from the same type of products.

According to the embodiment of the disclosure, the product recommendation is performed on the user by combining the historical data, so that the user has higher acceptance of the recommended product, and the product sales volume is increased through the repeated purchase of the user. In addition, the requirements of the user are mined through historical data, for example, the same type of products are recommended, and therefore more choices are provided for the user on the basis that the user has higher acceptance.

Fig. 5 schematically shows a flow diagram of a product recommendation method according to another embodiment of the present disclosure.

As shown in fig. 5, the product recommendation method of this embodiment includes operations S510 to S520, and S220 to S230.

In operation S510, the user is classified by the user classification model to obtain a loyalty classification.

According to an embodiment of the present disclosure, before operation S510, a user classification model may also be trained. Taking the first product as a financial product and the institution as a financial institution as an example, the details are as follows.

Firstly, a plurality of training users in a training set are obtained, and category labels are labeled for the training users. And obtaining interaction data of each user in the training set and an organization to which the N first products belong, for example, the interaction data includes: 1) identity information such as gender, age, marital status, work, etc. registered by the user at the institution. 2) The level information of the user in the organization, such as user visitors information, namely user star level, vip level, time registered as vip, and the like. 3) The user's asset information, such as account balance information, liability information, balance sheet, etc. 4) The historical data of products purchased by the user at the institution, such as the transaction amount of the products, the types of the products and the like.

In some embodiments of the present disclosure, the financial institution has characteristics of large number of users and various product types, the transaction data information (such as interactive data) newly added every day increases at an exponential rate, and the data occurs in different business systems and departments, and the newly added user data in each channel can be sent to a data warehouse for storage. The user data in the training set may be obtained from a data repository under conditions that comply with legal regulations.

Then, data preprocessing and normalization processing, such as processing of redundant data, are performed. Similar attributes of users may be presented in different forms, with high correlation between different presentation forms. By removing data redundancy, a more optimized training model is favorably established, and the truth of the user classification model is improved. The redundancy of the two samples can be compared by adopting a Pearson correlation coefficient detection method, and unnecessary data can be deleted.

Then, a bp (back propagation) neural network can be used to obtain the user classification model. The BP neural network may include an input layer, a hidden layer and an output layer. And inputting the processed interactive data serving as training characteristic data into a BP neural network, and outputting a classification result which is the same as the user class label, namely different loyalty classes through continuous training. In the training process, a back propagation algorithm can be used for updating parameter values, and a Sigmoid function is selected at an output layer to output a probability value so as to determine a classification result.

Finally, a plurality of test users in the test set are obtained, the trained user classification model is used for classifying the test users and obtaining test results, and if the test results are in accordance with expectations, the user classification model is used for executing operation S510.

In operation S520, each of the first products is classified by the product classification model to obtain a sales classification.

According to an embodiment of the present disclosure, before operation S520, a user classification model may also be trained. Taking the first product as a financial product and the institution as a financial institution as an example, the details are as follows.

First, a plurality of training products in a training set are obtained, and category labels are labeled for the training products. And obtains sales data for each product in the training set, e.g., 1) a sales channel, such as an online channel, an offline channel, or a sales channel of a partner facility, etc. 2) The promotion times, such as short message pushing times, offline advertisement laying times, telephone marketing times and the like in a certain time period. 3) Expected rate of return, such as expected rate of return for a period of time in the future. 4) Historical collection of vegetables, such as actual collection of vegetables over a period of time in the past. 5) And purchasing means such as offline agency purchase, partner agency purchase, online purchase and the like.

Then, data preprocessing and normalization processing are performed. Such as to reduce noisy data or to remove redundant data. The noise-reduced data may be processed for scrambling codes and nulls in the data.

Next, a product classification model can be obtained by using LSTM (Long Short-Term Memory) neural network training. And taking the processed sales data as training characteristic data, and outputting the data as the sales volume category of the first product.

Finally, a plurality of test products in the test set are obtained, the trained product classification model is used for classifying the test products and obtaining test results, and if the test results are in accordance with expectations, the product classification model is used for executing operation S520.

In operation S220, a corresponding evaluation index is obtained based on the sales amount category of each first product using a preset evaluation condition.

In operation S230, M second products are determined from the N first products based on the loyalty category to recommend to the user, wherein an evaluation index of each second product is greater than or equal to a preset threshold, and N and M are integers greater than or equal to 1, respectively.

Although the various operations of the methods are described above in a particular order, embodiments of the disclosure are not so limited, and the operations described above may be performed in other orders as desired. For example, operation S510 may be performed after operation S520, or may be performed simultaneously.

According to embodiments of the present disclosure, the number and quality of users of an organization are important factors in determining the viability and quality of development of an organization. Through accurate user classification, corresponding recommendation schemes can be preset for different types of users according to classification results, and the method is favorable for maintaining stable user relationship and further improving profit of mechanisms. Different products have different characteristics, different users are different, and the profits brought by the different products are different, so that the product classification model can input the classification results of the products to help the mechanism determine the product recommendation scheme, the mechanism profits can be improved, and more suitable products can be brought to the users.

Fig. 6 schematically shows a flowchart of obtaining loyalty categories in operation S510 according to an embodiment of the present disclosure.

As shown in fig. 6, classifying the user by the user classification model to acquire the loyalty program in operation S510 includes operations S610 to S620.

In operation S610, first feature data of a user is acquired, where the first feature data includes interaction data of the user with an organization to which the N first products belong.

According to the embodiment of the present disclosure, the type of the first feature data during training may be the same as that of the user classification model during use, wherein the content of the interactive data may refer to the training process for training the user classification model, which is not described herein again.

In operation S620, the first feature data is input to the user classification model to obtain the loyalty program.

According to the embodiment of the disclosure, compared with the method of classifying the users by questionnaires, purchasing product quota and the like, the trust degree of the users to the mechanism can be more comprehensively excavated through the interactive data, the trust degree can reflect the requirement degrees of the users to different products of the mechanism and the success rate of the users after recommending the products with pertinence, so that the users can be more objectively classified by using a trained user classification model, and the pertinence of recommending the products for the users is improved.

Fig. 7 schematically shows a flowchart of obtaining sales categories in operation S520 according to an embodiment of the present disclosure.

As shown in fig. 7, classifying each first product by the product classification model to obtain the sales category in operation S520 includes operations S710 to S720.

In operation S710, second characteristic data of each first product is acquired, wherein the second characteristic data includes sales data of each first product.

According to the embodiment of the present disclosure, the type of the second feature data during training may be the same as that of the product classification model during use, wherein the content of the sales data may refer to the training process of the product classification model, which is not described herein again.

In operation S720, the second characteristic data of each first product is input to the product classification model to predict a sales category corresponding to each first product.

The reason why the sales amount category can be predicted based on the characteristic data such as the sales channel, the promotion times, the expected profitability, the historical profit, the purchase mode and the like is, for example, that the number of users facing different sales channels is different and the degree of understanding of the users is different. More promotion times result in higher product exposure, but too many promotion times may cause user's dislike. The user may mutually certify the expected profitability as well as the historical profitability to decide whether to purchase the product. Also, for example, users of different ages may receive different purchases and may have different amounts of money at their disposal.

In the related art, the expected sales volume for the first product is generally predicted by a human. For example, if the expected sales volume of a financial product is predicted incorrectly, the credit distribution plan of the financial product, and even the annual profit plan of the institution, may be negatively impacted. According to the embodiment of the disclosure, the first product is classified into different sales levels by the product classification model through the sales data of the first product, so that an organization can make an accurate plan. In addition, the more sales, the more popular the market, so that the user can be recommended a better product by using the sales as at least one factor of the evaluation index.

Fig. 8 schematically shows a flowchart of obtaining the evaluation index in operation S220 according to an embodiment of the present disclosure.

As shown in fig. 8, obtaining the corresponding evaluation index based on the sales amount category of each first product using the preset evaluation condition in operation S220 includes operations S810 to S830.

In operation S810, feedback data for each first product is acquired, wherein the feedback data includes user evaluation data for each first product during a sales process.

In operation S820, a first weight is assigned to the feedback data, and a second weight is assigned to the sales amount category to obtain an evaluation index, wherein the evaluation condition includes the first weight and the second weight.

In some embodiments of the present disclosure, as the first products are promoted, feedback data such as recency, sales category, user evaluation data, etc. of each first product may be dynamically collected, for example, a weighted average of the above data is calculated, and the higher the product evaluation index is, the higher the product evaluation index is.

In some embodiments of the present disclosure, the evaluation condition may include setting a comparison table of different feedback data and different first weights in advance, and a comparison table of different sales categories and different second weights. Further, corresponding first and second weights may be determined based on the feedback data and the sales category, and then calculated, for example, by summing the first and second weights (for example only) to obtain an evaluation index.

In other embodiments of the present disclosure, the feedback data may be processed to map to a specific numerical value, such as a product rating score. And processes the sales categories to map to specific numerical values, e.g., higher sales and higher scores. And finally, calculating to obtain a first numerical value based on the value of the feedback data and the first weight, calculating to obtain a second numerical value based on the value of the sales and the second weight, and finally obtaining the evaluation index based on the first numerical value and the second numerical value. The specific calculation process may be direct addition or multiplication, or may be obtained based on functions such as an exponential function, a power function, or a gaussian function, and the disclosure is not particularly limited.

According to the embodiment of the disclosure, the evaluation indexes of the first products are obtained through calculation of multiple dimensions, and the characteristics of each first product can be reflected more scientifically. In addition, the evaluation index of each first product can be dynamically updated by dynamically acquiring the feedback data or sales volume category of each first product at intervals of a preset time period, so that the recommended products for users can be adjusted in time based on the wind direction of the current market, and the pertinence and the product sales profits of organizations are improved.

In an embodiment according to the present disclosure, after a user purchases a product, the product may be recommended to other users. Thus, the number of times recommended for each product can be tracked, and the feedback data can include the number of times recommended, thereby incorporating the number of times recommended into the evaluation conditions as one of the factors for obtaining the evaluation index.

Based on the product recommendation method, the disclosure also provides a product recommendation device. The apparatus will be described in detail below with reference to fig. 9.

Fig. 9 schematically shows a block diagram of a product recommendation device according to an embodiment of the present disclosure.

As shown in fig. 9, the product recommendation apparatus 900 of this embodiment includes a category acquisition module 910, an evaluation index module 920, and a product recommendation module 930.

The category obtaining module 910 may, for example, perform operation S210 to obtain a loyalty category corresponding to the user and a sales category of each of the N first products.

According to an embodiment of the present disclosure, the category acquisition module 910 may be configured to perform operation S510, and in particular, may be configured to perform operations S610 to S620. The loyalty classification method is used for classifying users through a user classification model to obtain loyalty classifications, and specifically comprises the following steps: first characteristic data of a user are obtained, wherein the first characteristic data comprise interaction data of the user and mechanisms to which the N first products belong. The first feature data is input to a user classification model to obtain a loyalty classification.

According to an embodiment of the present disclosure, the category acquisition module 910 may be configured to perform operation S520, and in particular, may be configured to perform operations S710 to S720. The method is used for classifying each first product through a product classification model to obtain a sales amount category, and specifically comprises the following steps: second characteristic data of each first product is obtained, wherein the second characteristic data comprises sales data of each first product. And inputting the second characteristic data of each first product into a product classification model to predict the sales amount category corresponding to each first product.

The evaluation index module 920 may perform, for example, operation S220 for obtaining a corresponding evaluation index based on the sales amount category of each first product using a preset evaluation condition.

The category acquisition module 910 may be configured to perform operations S810 to S820 according to an embodiment of the present disclosure. And the feedback data is used for acquiring the feedback data of each first product, wherein the feedback data comprises user evaluation data of each first product in the sale process. And allocating a first weight to the feedback data and allocating a second weight to the sales amount category to obtain an evaluation index, wherein the evaluation condition comprises the first weight and the second weight.

The product recommending module 930 may perform operation S230, for example, to determine M second products from the N first products based on the loyalty category for recommending to the user, where the evaluation index of each second product is greater than or equal to a preset threshold, and N and M are integers greater than or equal to 1, respectively.

The product recommendation module 930 may also perform operations S310 through S320, for example, according to an embodiment of the present disclosure. And the system is used for acquiring historical data of products purchased by the user under the condition of user authorization, and determining M second products based on the loyalty classification and the historical data under the condition of acquiring the historical data.

According to an embodiment of the present disclosure, the product recommendation module 930 may further perform operations S410 to S420, for example, where the operations S420 include operations S421 to S422. The method is used for determining S third products from historical data, wherein the N first products comprise the S third products, the evaluation index of each third product is larger than or equal to a preset threshold, and S is an integer larger than or equal to 1. Determining M second products based on the loyalty category, wherein: and taking the S third products as S second products, wherein S is less than or equal to M. Or based on the product types of the S third products, M first products of the same type are determined from the N first products to serve as M second products.

According to an embodiment of the present disclosure, any multiple modules of the category obtaining module 910, the evaluation index module 920 and the product recommending module 930 may be combined into one module to be implemented, or any one of the modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of the other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the category obtaining module 910, the evaluation index module 920, and the product recommending module 930 may be implemented at least partially as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented by hardware or firmware in any other reasonable manner of integrating or packaging a circuit, or implemented by any one of three implementations of software, hardware, and firmware, or any suitable combination of any of them. Alternatively, at least one of the category obtaining module 910, the evaluation index module 920 and the product recommendation module 930 may be at least partially implemented as a computer program module, which when executed, may perform a corresponding function.

FIG. 10 schematically illustrates a block diagram of an electronic device suitable for implementing a method of product recommendation in accordance with an embodiment of the present disclosure.

As shown in fig. 10, an electronic device 1000 according to an embodiment of the present disclosure includes a processor 1001 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)1002 or a program loaded from a storage section 1008 into a Random Access Memory (RAM) 1003. Processor 1001 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or associated chipset, and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), among others. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of a method flow according to embodiments of the present disclosure.

In the RAM 1003, various programs and data necessary for the operation of the electronic apparatus 1000 are stored. The processor 1001, ROM 1002, and RAM 1003 are connected to each other by a bus 1004. The processor 1001 performs various operations of the method flow according to the embodiments of the present disclosure by executing programs in the ROM 1002 and/or the RAM 1003. Note that the program may also be stored in one or more memories other than the ROM 1002 and the RAM 1003. The processor 1001 may also perform various operations of method flows according to embodiments of the present disclosure by executing programs stored in one or more memories.

Electronic device 1000 may also include an input/output (I/O) interface 1005, the input/output (I/O) interface 1005 also being connected to bus 1004, according to an embodiment of the present disclosure. Electronic device 1000 may also include one or more of the following components connected to I/O interface 1005: an input section 1006 including a keyboard, mouse, and the like. Including an output portion 1007 such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker and the like. A storage section 1008 including a hard disk and the like. And a communication section 1009 including a network interface card such as a LAN card, a modem, or the like. The communication section 1009 performs communication processing via a network such as the internet. The driver 1010 is also connected to the I/O interface 1005 as necessary. A removable medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 1010 as necessary, so that a computer program read out therefrom is mounted into the storage section 1008 as necessary.

The present disclosure also provides a computer-readable storage medium, which may be embodied in the devices/apparatuses/systems described in the above embodiments. Or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.

According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, a computer-readable storage medium may include the ROM 1002 and/or the RAM 1003 described above and/or one or more memories other than the ROM 1002 and the RAM 1003.

Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the method illustrated in the flow chart. The program code is for causing a computer system to carry out the method according to the embodiments of the disclosure, when the computer program product is run on the computer system.

The computer program performs the above-described functions defined in the system/apparatus of the embodiments of the present disclosure when executed by the processor 1001. The systems, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.

In one embodiment, the computer program may be hosted on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program may also be transmitted in the form of a signal on a network medium, distributed, downloaded and installed via the communication part 1009, and/or installed from the removable medium 1011. The computer program containing program code may be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

In such an embodiment, the computer program may be downloaded and installed from a network through the communication part 1009 and/or installed from the removable medium 1011. The computer program performs the above-described functions defined in the system of the embodiment of the present disclosure when executed by the processor 1001. The systems, devices, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.

In accordance with embodiments of the present disclosure, program code for executing computer programs provided by embodiments of the present disclosure may be written in any combination of one or more programming languages, and in particular, these computer programs may be implemented using high level procedural and/or object oriented programming languages, and/or assembly/machine languages. The programming language includes, but is not limited to, programming languages such as Java, C + +, python, the "C" language, or the like. The program code may execute entirely on the user computing device, partly on the user device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described separately above, this does not mean that the measures in the embodiments cannot be used in advantageous combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be devised by those skilled in the art without departing from the scope of the present disclosure, and such alternatives and modifications are intended to be within the scope of the present disclosure.

Claims (12)

1. A product recommendation method, comprising: Obtain the loyalty category corresponding to the user and the sales category of each first product in the N first products; Using a preset evaluation condition, obtain a corresponding evaluation index based on the sales category of each first product; Based on the loyalty category, M second products are determined from the N first products to recommend to the user, wherein the evaluation index of each second product is greater than or equal to a preset threshold, N and M are integers greater than or equal to 1, respectively. 2. The product recommendation method according to claim 1, wherein the determining M second products from the N first products based on the loyalty category to recommend to the user comprises: In the case of the user's authorization, obtain the historical data of the product purchased by the user; In the case of acquiring the historical data, the M second products are determined based on the loyalty category and the historical data. 3. The product recommendation method according to claim 2, wherein the determining the M second products based on the loyalty category and the historical data comprises: S third products are determined from the historical data, wherein the N first products include the S third products, and the evaluation index of each third product is greater than or equal to the preset threshold , S is an integer greater than or equal to 1; The M second products are determined based on the loyalty category, including: Using the S third products as the S second products, where S is less than or equal to M; or Based on the product types of the S third products, M first products of the same type are determined from the N first products as the M second products. 4. The product recommendation method according to claim 1, wherein the obtaining the loyalty category corresponding to the user comprises classifying the user through a user classification model to obtain the loyalty category, which specifically includes: acquiring first feature data of the user, wherein the first feature data includes interaction data between the user and the institutions to which the N first products belong; The loyalty category is obtained by inputting the first characteristic data into the user classification model. 5. The product recommendation method according to claim 4, wherein the interaction data comprises at least one of the following: The identity information of the user registered in the institution, the level information of the user in the institution, the asset information of the user, and the historical data of the product purchased by the user in the institution. 6. The product recommendation method of claim 1, wherein the obtaining the sales category of each first product comprises classifying the each first product by a product classification model to obtain the sales categories, which, in particular, include: acquiring second characteristic data of each first product, wherein the second characteristic data includes sales data of each first product; The second characteristic data of each first product is input into the product classification model to predict the sales category corresponding to each first product. 7. The product recommendation method according to claim 6, wherein the sales data comprises at least one of the following: of each first product, Sales channel, number of promotions, expected rate of return, historical income, purchase method. 8. The product recommendation method according to claim 6, wherein, using a preset evaluation condition to obtain a corresponding evaluation index based on the sales category of each first product comprises: Acquiring feedback data of each first product, wherein the feedback data includes user evaluation data of each first product in the sales process; The evaluation index is obtained by assigning a first weight to the feedback data and assigning a second weight to the sales category, wherein the evaluation condition includes the first weight and the second weight. 9. A product recommendation device, comprising: The category acquisition module is used to acquire the loyalty category corresponding to the user and the sales category of each first product in the N first products; an evaluation index module, configured to obtain a corresponding evaluation index based on the sales category of each first product by using a preset evaluation condition; A product recommendation module, configured to determine M second products from the N first products based on the loyalty category to recommend them to the user, wherein the evaluation index of each second product is greater than or equal to the preset threshold, N and M are integers greater than or equal to 1, respectively. 10. An electronic device comprising: one or more processors; storage means for storing one or more programs, Wherein, when the one or more programs are executed by the one or more processors, the one or more processors are caused to perform the method according to any one of claims 1-8. 11. A computer-readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method of any one of claims 1-8. 12. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-8.

Images