CN114693409A - Product matching method, device, computer equipment, storage medium and program product - Google Patents

Product matching method, device, computer equipment, storage medium and program product Download PDF

Info

Publication number
CN114693409A
CN114693409A CN202210445601.8A CN202210445601A CN114693409A CN 114693409 A CN114693409 A CN 114693409A CN 202210445601 A CN202210445601 A CN 202210445601A CN 114693409 A CN114693409 A CN 114693409A
Authority
CN
China
Prior art keywords
user
product
financial
grade
historical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210445601.8A
Other languages
Chinese (zh)
Inventor
贾颖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Industrial and Commercial Bank of China Ltd ICBC
Original Assignee
Industrial and Commercial Bank of China Ltd ICBC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Industrial and Commercial Bank of China Ltd ICBC filed Critical Industrial and Commercial Bank of China Ltd ICBC
Priority to CN202210445601.8A priority Critical patent/CN114693409A/en
Publication of CN114693409A publication Critical patent/CN114693409A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0623Item investigation
    • G06Q30/0625Directed, with specific intent or strategy
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0631Item recommendations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/02Banking, e.g. interest calculation or account maintenance
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/06Asset management; Financial planning or analysis

Abstract

The application relates to a product matching method, a product matching device, a computer device, a storage medium and a computer program product. Acquiring historical behavior data and account balance data of a user, and inputting the historical behavior data and the account balance data into a preset classification model to obtain a user grade for representing the financial risk resistance of the user; determining a first target product matched with the user from a product database according to the user grade; the historical behavior data comprises historical search records and historical browsing records of a user on financial products, and the product database comprises a plurality of financial products with tag information; the financial risk resistance level of the user is determined by analyzing the historical search records and the historical browsing records of the financial products of the user and analyzing the historical account balance data of the user, so that more appropriate financial products are matched for the user according to the financial risk resistance level of the user, and the matching degree of the financial products and the user is improved.

Description

Product matching method, device, computer equipment, storage medium and program product
Technical Field
The present application relates to the field of computer technologies, and in particular, to a product matching method, apparatus, computer device, storage medium, and computer program product.
Background
In the financial industry, in order to meet the diversified financial demand of users, different types of financial products are increasing, for example, different types of financial products and different types of deposit products set by banks according to different demands of users, and how to efficiently promote the financial products to the customers is an important issue to be solved at present.
In the traditional technology, when financial products are promoted, different financial products are promoted for a user through counter marketing personnel of a bank when the user walks into a bank business outlet; alternatively, different financial products are pushed to the user through the bank's client (i.e., the bank's client on the user's terminal).
However, the existing financial product marketing method is not targeted, and the financial product which is marketed to the user has a problem of low matching degree with the user.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a product matching method, apparatus, computer device, computer readable storage medium, and computer program product capable of improving the matching degree of a product and a user.
In a first aspect, the present application provides a product matching method. The method comprises the following steps:
acquiring historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the financial products;
inputting historical behavior data and account balance data into a preset classification model to obtain a user grade of a user; the user grade is used for representing the financial risk resistance of the user;
determining a first target product matched with the user from a product database according to the user grade; wherein the product database includes a plurality of financial products having tag information.
In one embodiment, the step of inputting the historical behavior data and the account balance data into a preset classification model to obtain the user grade of the user comprises the following steps:
acquiring a feature vector of a user; the feature vector of the user comprises a first feature vector generated based on historical behavior data and a second feature vector formed based on account balance data;
inputting the feature vector into a preset classification model to obtain the user grade of the user; the preset classification model is obtained by training an initial classification model by adopting a plurality of sample feature vectors and corresponding grade labels.
In one embodiment, obtaining a feature vector of a user includes:
extracting the features of the historical search records to obtain a first feature subvector; the characteristic value in the first characteristic sub-vector comprises search time, search equipment and a search keyword;
extracting the features of the historical browsing records to obtain a second feature sub-vector; the characteristic values in the second characteristic sub-vector comprise browsing time, browsing equipment and historical browsing product types; the first feature vector comprises a first feature sub-vector and a second feature sub-vector;
obtaining a second feature vector based on the account balance data; the characteristic values in the second characteristic vector comprise total income amount, total expenditure amount and minimum income amount per month;
and performing feature fusion on the first feature sub-vector, the second feature sub-vector and the second feature vector to obtain a feature vector of the user.
In one embodiment, determining a first target product matching the user from the product database according to the user rating comprises:
acquiring label information of each financial product in a product database; the tag information of the financial product includes at least one of a product type, a purchase standard, a risk level, an income level, and an age;
and determining a first target product matched with the user according to the user grade and the label information of each financial product.
In one embodiment, determining a first target product matching the user according to the user grade and the tag information of each financial product comprises:
determining a reference level corresponding to each label information of the financial product aiming at each financial product in a product database, and determining a user matching degree corresponding to the financial product according to the user level and the reference level corresponding to each label information;
and selecting a first target product matched with the user according to the user matching degree corresponding to each financial product.
In one embodiment, determining the user matching degree corresponding to the financial product according to the user grade and the reference grade corresponding to each tag information includes:
determining target label information matched with the user grade in each label information of the financial product according to the reference grade corresponding to each label information;
and performing weighting processing on the target label information to determine the user matching degree corresponding to the financial product.
In one embodiment, selecting a first target product matched with a user according to the user matching degree corresponding to each financial product comprises:
sorting the user matching degrees corresponding to the financial products according to the descending order;
and taking the financial products with the pre-set number after the sorting processing as first target products matched with the user.
In one embodiment, the method further comprises:
acquiring current search information input by a user through a target client;
determining a second target product matched with the current search information and the user grade from a product database according to the current search information and the user grade;
and sending the second target product to the target client.
In a second aspect, the application also provides a product matching device. The device includes:
the first acquisition module is used for acquiring historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the financial products;
the second acquisition module is used for inputting the historical behavior data and the account balance data into a preset classification model to acquire the user grade of the user; the user grade is used for representing the financial risk resistance of the user;
the determining module is used for determining a first target product matched with the user from the product database according to the user grade; wherein the product database includes a plurality of financial products having tag information.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:
acquiring historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the financial products;
inputting historical behavior data and account balance data into a preset classification model to obtain a user grade of a user; the user grade is used for representing the financial risk resistance of the user;
determining a first target product matched with the user from a product database according to the user grade; wherein, the product database comprises a plurality of financial products with label information.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
acquiring historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the financial products;
inputting historical behavior data and account balance data into a preset classification model to obtain a user grade of a user; the user grade is used for representing the financial risk resistance of the user;
determining a first target product matched with the user from a product database according to the user grade; wherein the product database includes a plurality of financial products having tag information.
In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of:
acquiring historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the financial products;
inputting historical behavior data and account balance data into a preset classification model to obtain a user grade of a user; the user grade is used for representing the financial risk resistance of the user;
determining a first target product matched with the user from a product database according to the user grade; wherein the product database includes a plurality of financial products having tag information.
According to the product matching method, the product matching device, the computer equipment, the storage medium and the computer program product, the user grade for representing the financial risk resistance of the user is obtained by acquiring the historical behavior data and the account balance data of the user and inputting the historical behavior data and the account balance data into the preset classification model; then, according to the user grade, determining a first target product matched with the user from a product database; the historical behavior data comprises historical search records and historical browsing records of a user on financial products, and the product database comprises a plurality of financial products with tag information; that is to say, the product matching method provided by the embodiment of the application determines the anti-financial risk level of the user by analyzing the historical search record and the historical browsing record of the financial product related to the user and analyzing the historical account balance data of the user, and further matches the more appropriate financial product for the user according to the anti-financial risk level of the user; the financial products are matched by comprehensively evaluating the user preference and the financial risk resistance, so that the matching degree of the financial products and the user can be greatly improved, more accurate personalized customized financial services are provided for the user, financial products with stronger pertinence are provided for different users, the satisfaction degree of the user can be improved, and the efficiency of financial product recommendation is improved.
Drawings
FIG. 1 is a diagram of an application environment of a product matching method in one embodiment;
FIG. 2 is a schematic flow chart diagram illustrating a product matching method in one embodiment;
FIG. 3 is a schematic flow chart diagram of a product matching method in another embodiment;
FIG. 4 is a schematic flow chart diagram of a product matching method in another embodiment;
FIG. 5 is a schematic flow chart diagram of a product matching method in another embodiment;
FIG. 6 is a schematic flow chart diagram of a product matching method in another embodiment;
FIG. 7 is a block diagram showing the structure of a product matching apparatus according to an embodiment;
FIG. 8 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.
First, before specifically describing the technical solution of the embodiment of the present disclosure, a technical background or a technical evolution context on which the embodiment of the present disclosure is based is described. Generally, when a financial product is promoted, two ways may be included, one of which is to send a recommended financial product to a user through a bank client (such as a client application program of a mobile phone bank, an internet bank, etc.); in this way, the demand of different users for financial products is different due to the high and low risk resistance and income difference, and simple random promotion or profit-oriented financial product promotion can not only reduce the satisfaction degree of users, but also cause the loss of user benefits and is not beneficial to the long-term marketing business development. In summary, inaccurate positioning of the financial product results in a low degree of match between the financial product and the user.
And secondly, counter marketing is carried out through business outlets, product marketing of the business outlets is usually carried out through tellers or hall managers, the cognition and risk control of financial products by financial practitioners are limited, product explanation and the establishment of an overall financing scheme are difficult to be provided for users in detail, and a large amount of manpower and material resources are needed to support financial product marketing. In summary, the financial business personnel have a weak relationship between the financial product and the user, resulting in a low degree of matching between the financial product and the user.
Based on this, the embodiment of the application provides a product matching method, which can perform deep analysis and positioning on a user through historical behavior data and account balance data of the user to determine a user grade, perform label refinement on a financial product, and further determine a financial product with a higher matching degree with the user according to the user grade and label information of the financial product, so as to improve the matching degree between the financial product and the user. In addition, accurate recommendation of financial products is carried out by analyzing interests, risk carrying capacity, income and expenditure conditions and the like of the users, so that a personalized and highly-targeted financial management scheme can be provided for the users, the business pressure of marketing personnel is reduced, the satisfaction degree of the users is improved, and the effectiveness and the accuracy of financial product marketing can be effectively improved.
The following describes technical solutions related to the embodiments of the present application with reference to a scenario in which the embodiments of the present application are applied.
The product matching method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104, or may be located on the cloud or other network server. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, portable wearable devices, and intelligent operation terminals in financial business halls, and the like, the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart car-mounted devices, and the like, and the portable wearable devices may be smart watches, smart bracelets, head-mounted devices, and the like; a client application for performing a financial operation may be installed in the terminal 102. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers. The user accesses the client application program through the terminal 102, and performs financial business operation through the client application program, and the server 104 matches a more appropriate financial product for the user through the historical behavior data of the user and the historical account balance data of the user, and pushes the more appropriate financial product to the user through the client application program.
In one embodiment, as shown in fig. 2, a product matching method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:
step 201, obtaining historical behavior data and account balance data of a user.
Where the historical behavioral data includes historical search records and historical browsing records of the user for financial products, the account balance data may include, but is not limited to, transaction data for all revenues and payouts of the user's account, and credit transaction data for the user.
Optionally, the user logs in the client application program through a login account, and after entering the client application program, financial operations related to financial services may be performed, such as: browsing financial products, searching financial products, sharing financial products, clicking to access financial products, performing financial transfer operation and the like; by monitoring the behavior operation of the user in the client application program, the historical search record and the historical browsing record of the user can be acquired; optionally, for real-time data streaming, data collection may be performed by using a Flume distributed log aggregation system monitoring file, log files related to the user are automatically collected by performing file configuration, and the log files are transmitted to Kafka; accessing historical behaviors (including but not limited to logging, browsing, searching, sharing, clicking and other behaviors) of a user and data sets such as system running logs, and analyzing and processing the data in real time by using a message queue system of the user to analyze the logs; using Flume to monitor user behaviors, and automatically acquiring file data and transmitting the file data to Kafka for processing when the behaviors are changed; so as to obtain the historical search record and the historical browsing record of the financial product of the user. Optionally, in order to improve data processing efficiency, an offline processing mode may be adopted, so that collected file data may be uniformly uploaded to a distributed file transmission system in a format for convenience of processing, and offline file processing is performed by using spark technology and stored in the HBase database.
Optionally, the server may obtain historical balance data of the user account as account balance data of the user, where the account balance data may include balance data of the user within a historical preset time period, or all balance data of the user in the history; the preset time period may be a month, a season, a year, etc.
Step 202, inputting the historical behavior data and the account balance data into a preset classification model to obtain the user grade of the user.
Wherein the user rating is used to characterize the user's financial risk resistance. Optionally, the user rating may be classified into three (e.g., low, medium, high), five (e.g., extremely low, medium, high), or any multiple of ratings, where a higher rating indicates a higher financial risk resistance of the user; in practical application, the user grade division can be flexibly adjusted and set, and the method is not limited in the application.
Optionally, the preset classification model may be a deep learning-based classification model, a neural network-based classification model, a machine learning-based classification model, or the like; for example: the preset classification model can classify users by utilizing a multi-layer perceptron classifier based on a feedforward neural network, a sigmod activation function is used by an intermediate node, a softmax function is used by an output layer for normalization processing, and nodes of the output layer represent the types of the classifier, namely different user grades of the users.
Through the preset classification model, the user grade can be classified according to the historical behavior data and the account balance data of the user, and the user grade matched with the user is determined; optionally, historical behavior data of the user and account balance data of the user may be input to the preset classification model, so as to obtain a user level corresponding to the user.
Step 203, according to the user grade, determining a first target product matched with the user from the product database.
The product database includes a plurality of financial products with tag information, that is, each financial product in the product database has different tag information, each financial product may correspond to at least one tag information, and the tag information may include, but is not limited to, at least one of a product type, a purchase standard of the financial product, a risk level of the financial product, an income level of the financial product, and an age of the financial product.
Alternatively, the server may determine a financial product matching the user grade as the first target product matching the user according to the user grade, the tag information of each financial product in the product database, and the preset corresponding relationship between different tags and the user grade. Alternatively, the financial products matching the user grade may include financial products of the same grade as the user or financial products of a lower grade than the user. In addition, in the case where the financial product includes a plurality of tag information, the financial product matching the user rank may be a financial product in which at least one tag information of the financial product matches the user rank, or a financial product in which a preset number of tag information of the plurality of tag information of the financial product matches the user rank exists, or a financial product in which a preset number of tag information of the plurality of tag information of the financial product matches the user rank, or the like.
For the above mentioned correspondence between different preset tags and user classes, it can be understood that each tag information may include a plurality of sub-tag information matched with different user classes, such as: the buy-up criteria can include buy-up criteria corresponding to a user rating of low financial risk resistance, buy-up criteria corresponding to a user rating of medium financial risk resistance, and buy-up criteria corresponding to a user rating of high financial risk resistance. Through the corresponding relation, whether certain label information of the financial product is matched with the user grade can be judged.
In addition, after the server determines a first target product matched with the user, the server can also push the first target product to the user; optionally, the first target product may be sent to a client application of the user (e.g., personal internet banking); the recommendation may also be performed through the service network, for example, the first target product is sent to a terminal device of the service network, so that a worker of the service network may recommend the first target product to the user when the user performs on-site service handling, or the worker of the service network may recommend the first target product to the user by using a voice call. It should be noted that, in the embodiment of the present application, a recommendation manner of a product is not specifically limited.
In the product matching method, the server acquires the user grade for representing the financial risk resistance of the user by acquiring the historical behavior data and the account balance data of the user and inputting the historical behavior data and the account balance data into a preset classification model; then, according to the user grade, determining a first target product matched with the user from a product database; the historical behavior data comprises historical search records and historical browsing records of a user on financial products, and the product database comprises a plurality of financial products with tag information; that is to say, the product matching method provided in the embodiment of the present application determines the financial risk resistance level of the user by analyzing the historical search records and the historical browsing records of the financial products related to the user and analyzing the historical account balance data of the user, and further matches the more appropriate financial products for the user according to the financial risk resistance level of the user; the financial products are matched by comprehensively evaluating the user preference and the financial risk resistance, so that the matching degree of the financial products and the user can be greatly improved, more accurate personalized customized financial services are provided for the user, financial products with stronger pertinence are provided for different users, the satisfaction degree of the user can be improved, and the efficiency of financial product recommendation is improved.
Fig. 3 is a flowchart illustrating a product matching method according to another embodiment. The present embodiment relates to an optional implementation process for inputting historical behavior data and account balance data into a preset classification model to obtain a user rating of a user, and based on the above embodiments, as shown in fig. 3, the step 202 includes:
step 301, obtaining a feature vector of a user.
Wherein the feature vector of the user comprises a first feature vector generated based on historical behavior data and a second feature vector formed based on account balance data.
Specifically, after the historical behavior data of the user is obtained, the historical behavior data of the user can be analyzed, and a first feature vector for representing the historical behavior features of the user is extracted; the first feature vector may include a first feature sub-vector and a second feature sub-vector, the first feature sub-vector is a feature vector generated based on the historical search record and used for characterizing the user search behavior, and the second feature sub-vector is a feature vector generated based on the historical browsing record and used for characterizing the user browsing behavior.
Optionally, feature extraction may be performed on a historical search record in the historical behavior data to obtain a first feature sub-vector for characterizing a user search behavior feature; wherein, the feature value in the first feature sub-vector may include but is not limited to search time, search device, search keyword, and the like; feature extraction can be carried out on historical browsing records in the historical behavior data to obtain a second feature sub-vector for representing the user browsing behavior feature; wherein, the feature values in the second feature sub-vector include but are not limited to browsing time, browsing device and historical browsing product type.
Optionally, for a plurality of historical search records of the user, clustering analysis may be performed on the plurality of historical search records by using clustering operation, and a first feature sub-vector for characterizing a user search behavior feature is generated according to a clustering result; such as: clustering analysis can be performed on the plurality of historical search records by adopting an LDA topic clustering algorithm to obtain the first characteristic subvector. When clustering operation is carried out, for each historical search record, a NLPIR Chinese word segmentation system can be adopted to carry out keyword extraction and part-of-speech tagging on each historical search record to obtain keyword feature vectors respectively corresponding to each historical search record, and then clustering analysis is carried out on the keyword feature vectors of each historical search record to obtain a first feature sub-vector corresponding to a user search behavior. Optionally, when extracting keywords, some words with insufficient characteristics may be further filtered, so as to extract a keyword feature vector that better reflects personal search characteristics of the user.
Similarly, for a plurality of historical browsing records of the user, the second feature sub-vector corresponding to the browsing behavior of the user can be obtained by adopting the same analysis method as the historical search record, and the specific process is not repeated in detail.
In addition, after the server acquires the account balance data of the user, the server can analyze the account balance data of the user and extract a second feature vector for representing the account balance behavior feature of the user; the feature values in the second feature vector include, but are not limited to, total income, total expenditure, minimum income per month, minimum expenditure per month, maximum income per month, maximum expenditure per month, and the like.
Further, after obtaining a first feature sub-vector for characterizing a user search behavior feature, a second feature sub-vector for characterizing a user browsing behavior feature, and a second feature vector for characterizing a user account balance behavior feature, the first feature sub-vector, the second feature sub-vector, and the second feature vector may be subjected to feature fusion to obtain a feature vector of the user. Optionally, the first feature sub-vector, the second feature sub-vector, and the second feature vector may be linearly transformed to a specific interval, so as to obtain a feature vector of the user. The redundancy processing may be performed on the first feature sub-vector, the second feature sub-vector, and the second feature vector, for example: replacing and updating some approximate words in each feature vector; then, carrying out linear transformation on the first feature sub-vector after redundancy processing, the second feature sub-vector after redundancy processing and the second feature vector after redundancy processing to obtain a feature vector of a user; the problem that the processing amount of the model is large and the accuracy of the output result of the model is reduced due to overlarge dimensionality of the feature vector can be effectively solved through the feature vector after redundant processing.
Step 302, inputting the feature vector to a preset classification model, and obtaining a user grade of the user.
The preset classification model is obtained by training an initial classification model by adopting a plurality of sample feature vectors and corresponding grade labels.
Optionally, the server may train the initial classification model in a supervised training manner according to the plurality of sample feature vectors and the level label corresponding to each sample feature vector, so as to obtain the preset classification model. Of course, in order to improve the accuracy of the classification model, some sample feature vectors may be constructed based on the historical behavior data and/or the account revenue and expenditure data of the user to add more sample data; then, based on the added sample feature vectors and the pre-collected sample feature vectors, the initial classification model is trained in a semi-supervised training mode to obtain the preset classification model.
Further, the server may determine the user level of the user through the preset classification model, that is, input the feature vector of the user to the preset classification model, and obtain the user level of the user.
In the embodiment, the user grade of the user is obtained by obtaining the feature vector of the user, including a first feature vector generated based on historical behavior data and a second feature vector formed based on account balance data, and inputting the feature vector into a preset classification model; the preset classification model is obtained by training an initial classification model by adopting a plurality of sample characteristic vectors and corresponding grade labels thereof; the first feature vector can represent the searching behavior and the browsing behavior of the user, and the second feature vector can represent the account balance behavior of the user, so that the user grade is determined based on the first feature vector and the second feature vector, the input data volume of the classification model can be reduced, the processing efficiency of the classification model is improved, the output accuracy of the classification model can be improved, and the accuracy of the user grade can be improved.
Fig. 4 is a flowchart illustrating a product matching method according to another embodiment. This embodiment relates to an alternative implementation process of determining a first target product matching a user from a product database according to a user rating, and based on the above embodiment, as shown in fig. 4, the step 203 includes:
step 401, obtaining label information of each financial product in the product database.
Wherein the tag information of the financial product includes at least one of a product type, purchase criteria, a risk level, an income level, and an age.
Alternatively, for the tag information of each financial product, the tag information of the financial product may be obtained by analyzing the related information of the financial product; the related information of the financial product may include, but is not limited to, the name, type, purchase range, income rate, annual interest rate, risk index, return rate, etc. of the financial product, and the tag value corresponding to each tag information of the financial product is determined by analyzing the related information of the financial product and the preset division rules of different tags; for example: the purchasing standard corresponding to the financial product can be determined according to the reference purchasing range and the actual purchasing range of the financial product respectively corresponding to different purchasing standards.
In one optional implementation manner of the embodiment, the server may classify financial products and deposit products that the bank needs to market; the financial products are classified according to the business, classified according to the risk and income, and different types of different ages of deposit products are set and classified at the same time, so that at least one piece of label information of each financial product is obtained, and a product database of the financial products is established.
Step 402, determining a first target product matched with the user according to the user grade and the label information of each financial product.
Optionally, for each piece of tag information of the financial product, since each piece of tag information may correspond to a different tag value, different levels may be preset for different tag values of one piece of tag information; the first purchase criteria and the second purchase criteria may be included as the purchase criteria, wherein the first purchase criteria is smaller than the second purchase criteria, a first rank may be set for the first purchase criteria, a second rank may be set for the second purchase criteria, and the first rank may be lower than the second rank.
Then, based on the tag information and the levels corresponding to the tag information, optionally, for each financial product in the product database, a reference level corresponding to each tag information of the financial product may be determined, and then, a user matching degree corresponding to the financial product may be determined according to the user level and the reference level corresponding to each tag information; furthermore, the first target product matched with the user can be selected according to the user matching degree corresponding to each financial product. That is to say, for each financial product, the user grade and the reference grade corresponding to each tag information of the financial product may be compared respectively to obtain the user matching degree corresponding to the financial product, optionally, in each tag information of the financial product, the more tag information matched with the user grade, the higher the user matching degree corresponding to the financial product is, that is, the user matching degree is positively correlated with the number of tag information matched with the user grade.
Further, after the user matching degrees corresponding to the financial products in the product database are obtained, the financial product with the higher user matching degree may be determined as the first target product matched with the user. Optionally, a financial product with a user matching degree greater than or equal to a preset matching degree threshold value can be determined as a first target product matched with the user; or sorting the user matching degrees corresponding to the financial products in descending order, and taking the financial products of the preset number before the sorting as the first target products matched with the users. In this embodiment, the determination manner of the first target product is not specifically limited.
In the embodiment, the server determines a first target product matched with the user according to the user level and the label information of each financial product by acquiring the label information of each financial product in the product database; wherein the tag information of the financial product includes at least one of a product type, a purchase criteria, a risk level, an income level, and an age; the user grade and the label information of the financial products are matched one by one to obtain a first target product with high matching degree with the user, so that the matching degree between the product and the user can be improved, and the product with high matching degree with the user and strong pertinence is obtained.
Fig. 5 is a flowchart illustrating a product matching method according to another embodiment. The present embodiment relates to an optional implementation process for determining a user matching degree corresponding to a financial product according to a user level and a reference level corresponding to each tag information, and on the basis of the foregoing embodiment, as shown in fig. 5, the foregoing method further includes:
step 501, determining target label information matched with the user grade in each piece of label information of the financial product according to the reference grade corresponding to each piece of label information.
Optionally, matching with the user level may be that the reference level corresponding to the tag information is the same as the user level, or that the reference level corresponding to the tag information is lower than the user level, that is, when the reference level corresponding to the tag information is the same as the user level, or the reference level corresponding to the tag information is lower than the user level, the tag information may be used as target tag information matching with the user level.
And 502, performing weighting processing on each target label information to determine the user matching degree corresponding to the financial product.
Optionally, different priority levels may be preset for a plurality of tag information of the financial product, that is, different weighting coefficients may be set for different tag information, where a higher weighting indicates a higher importance degree of the tag information, and may also indicate a higher association degree with the user; of course, the same weighting factor may be set for different tag information. Such as: the weight coefficient corresponding to the risk level in the tag information of the financial product may be higher than the weight coefficient corresponding to the purchase standard, and the weight coefficient corresponding to the risk level and the weight coefficient corresponding to the profit level may also be the same.
Based on this, after the target tag information of the financial product is determined, the weighting processing may be performed on each target tag information based on each target tag information and the weighting coefficient corresponding to each target tag information, so as to obtain the user matching degree corresponding to the financial product.
In the embodiment, the server determines target label information matched with the user grade in each piece of label information of the financial product according to the reference grade corresponding to each piece of label information, performs weighting processing on each piece of target label information, and determines the user matching degree corresponding to the financial product; different weight coefficients are set for different label information, so that the matching degree of the user corresponding to the financial product can represent the matching degree between the user and the product, the financial product with stronger pertinence is obtained, and the matching degree between the financial product and the user is improved.
Fig. 6 is a flowchart illustrating a product matching method according to another embodiment. The present embodiment relates to an optional implementation process for matching financial products for a user according to search information of the user, and based on the foregoing embodiment, as shown in fig. 6, the foregoing method further includes:
step 601, obtaining the current search information input by the user through the target client.
The current search information may be search information related to the financial product, including but not limited to product name, product type, purchase criteria, risk level, income level, age, and the like.
Alternatively, the target client may be a client at which the user logs into the client application through the user's login account.
Step 602, according to the current search information and the user level, determining a second target product matched with the current search information and the user level from the product database.
Optionally, the server may first screen out a candidate financial product corresponding to the current search information from the product database according to the current search information, and then determine a financial product matching the user level from the candidate financial products according to the user level as a second target product matching the user; the implementation process of determining, from the candidate financial products, the financial product that matches the user rank according to the user rank may refer to the relevant description process in the embodiments given in fig. 2 and fig. 4, and is not described herein again.
Optionally, the server may also screen candidate financial products matching the user rating from the product database according to the user rating, and then determine a financial product matching the current search information from the candidate financial products according to the current search information, as a second target product matching the user.
Step 603, sending the second target product to the target client.
In the embodiment, the server determines a second target product matched with the current search information and the user grade from the product database by acquiring the current search information input by the user through the target client and according to the current search information and the user grade; finally, sending the second target product to the target client; that is to say, in this embodiment, the server may further match a more suitable financial product for the user according to the search information of the user in combination with the user rank of the user, and compared with the prior art in which a financial product matched with the user is directly confirmed according to the search information of the user, in this embodiment, the matching degree between the user and the product can be improved by combining the user rank of the user.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the application also provides a product matching device for realizing the product matching method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so specific limitations in one or more embodiments of the product matching device provided below can be referred to the limitations of the product matching method in the foregoing, and details are not described herein again.
In one embodiment, as shown in fig. 7, there is provided a product matching device including: a first obtaining module 701, a second obtaining module 702, and a determining module 703, wherein:
a first obtaining module 701, configured to obtain historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the financial products;
a second obtaining module 702, configured to input the historical behavior data and the account balance data into a preset classification model, so as to obtain a user level of the user; the user grade is used for representing the financial risk resistance of the user;
a determining module 703, configured to determine, according to the user level, a first target product matched with the user from a product database; wherein the product database includes a plurality of financial products having tag information.
In one embodiment, the second obtaining module 702 includes a first obtaining unit and a second obtaining unit; the first acquisition unit is used for acquiring a feature vector of a user; the feature vector of the user comprises a first feature vector generated based on historical behavior data and a second feature vector formed based on account balance data; the second obtaining unit is used for inputting the feature vector to a preset classification model to obtain the user grade of the user; the preset classification model is obtained by training an initial classification model by adopting a plurality of sample feature vectors and corresponding grade labels.
In one embodiment, the first obtaining unit is specifically configured to perform feature extraction on a historical search record to obtain a first feature sub-vector; the characteristic value in the first characteristic sub-vector comprises search time, search equipment and a search keyword; extracting the features of the historical browsing records to obtain a second feature subvector; the characteristic values in the second characteristic sub-vector comprise browsing time, browsing equipment and historical browsing product types; the first feature vector comprises a first feature sub-vector and a second feature sub-vector; and obtaining a second feature vector based on the account balance data; the characteristic values in the second characteristic vector comprise total income amount, total expenditure amount and minimum income amount per month; and then, performing feature fusion on the first feature sub-vector, the second feature sub-vector and the second feature vector to obtain a feature vector of the user.
In one embodiment, the determining module 703 includes a third obtaining unit and a determining unit; the third acquisition unit is used for acquiring the label information of each financial product in the product database; the tag information of the financial product includes at least one of a product type, a purchase standard, a risk level, an income level, and an age; and the determining unit is used for determining a first target product matched with the user according to the user grade and the label information of each financial product.
In one embodiment, the determining unit is specifically configured to determine, for each financial product in the product database, a reference level corresponding to each tag information of the financial product, and determine, according to the user level and the reference level corresponding to each tag information, a user matching degree corresponding to the financial product; and selecting a first target product matched with the user according to the user matching degree corresponding to each financial product.
In one embodiment, the determining unit is specifically configured to determine, according to a reference level corresponding to each piece of tag information, target tag information that matches a user level in each piece of tag information of the financial product; and performing weighting processing on the target label information to determine the user matching degree corresponding to the financial product.
In one embodiment, the determining unit is specifically configured to sort the user matching degrees corresponding to the financial products in descending order; and taking the financial products with the pre-preset number after the sorting processing as first target products matched with the user.
In one embodiment, the apparatus further comprises a third obtaining module and a sending module; the third acquisition module is used for acquiring the current search information input by the user through the target client; the determining module 703 is further configured to determine, according to the current search information and the user rating, a second target product that matches the current search information and the user rating from the product database; and the sending module is used for sending the second target product to the target client.
The various modules in the product matching apparatus described above may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 8. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing the product database and the label information of each financial product in the product database, as well as historical behavior data of the user and account balance data of the user. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a product matching method.
Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the product matching method provided in the above embodiments when executing the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the product matching method provided in the respective embodiments described above.
In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the product matching method provided in the various embodiments described above.
It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases involved in the embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims (12)

1. A method of product matching, the method comprising:
acquiring historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the user for financial products;
inputting the historical behavior data and the account balance data into a preset classification model to obtain the user grade of the user; the user grade is used for characterizing the financial risk resistance of the user;
determining a first target product matched with the user from a product database according to the user grade; wherein the product database includes a plurality of financial products having tag information.
2. The method of claim 1, wherein the inputting the historical behavior data and the account balance data into a preset classification model to obtain the user rating of the user comprises:
acquiring a feature vector of the user; the feature vector of the user comprises a first feature vector generated based on the historical behavior data and a second feature vector formed based on the account balance data;
inputting the feature vector to the preset classification model to obtain the user grade of the user; the preset classification model is obtained by training an initial classification model by adopting a plurality of sample feature vectors and corresponding grade labels.
3. The method according to claim 2, wherein the obtaining the feature vector of the user comprises:
extracting the characteristics of the historical search records to obtain a first characteristic sub-vector; the characteristic values in the first characteristic sub-vector comprise search time, search equipment and search keywords;
extracting features of the historical browsing records to obtain a second feature sub-vector; the characteristic values in the second characteristic sub-vector comprise browsing time, browsing equipment and historical browsing product types; the first feature vector comprises the first feature sub-vector and the second feature sub-vector;
obtaining the second feature vector based on the account balance data; the characteristic values in the second characteristic vector comprise total income amount, total expenditure amount and minimum income amount per month;
and performing feature fusion on the first feature sub-vector, the second feature sub-vector and the second feature vector to obtain the feature vector of the user.
4. The method of any one of claims 1 to 3, wherein the determining a first target product matching the user from a product database according to the user rating comprises:
acquiring label information of each financial product in the product database; the tag information of the financial product includes at least one of a product type, a purchase standard, a risk level, an income level, and an age;
and determining a first target product matched with the user according to the user grade and the label information of each financial product.
5. The method of claim 4, wherein determining the first target product matching the user based on the user rating and the tag information of each of the financial products comprises:
for each financial product in the product database, determining a reference grade corresponding to each label information of the financial product, and determining a user matching degree corresponding to the financial product according to the user grade and the reference grade corresponding to each label information;
and selecting a first target product matched with the user according to the user matching degree corresponding to each financial product.
6. The method of claim 5, wherein the determining the user matching degree corresponding to the financial product according to the reference level corresponding to the user level and each tag information comprises:
determining target label information matched with the user grade in each piece of label information of the financial product according to the reference grade corresponding to each piece of label information;
and performing weighting processing on the target label information to determine the user matching degree corresponding to the financial product.
7. The method of claim 5, wherein selecting the first target product matching the user according to the user matching degree corresponding to each financial product comprises:
sorting the user matching degrees corresponding to the financial products according to the descending order;
and taking the financial products of the top preset number after the sorting processing as first target products matched with the user.
8. The method of claim 1, further comprising:
acquiring current search information input by the user through a target client;
according to the current search information and the user grade, determining a second target product matched with the current search information and the user grade from the product database;
and sending the second target product to the target client.
9. A product matching apparatus, the apparatus comprising:
the first acquisition module is used for acquiring historical behavior data and account balance data of a user; the historical behavior data comprises historical search records and historical browsing records of the financial products of the user;
the second acquisition module is used for inputting the historical behavior data and the account balance data into a preset classification model to acquire the user grade of the user; the user grade is used for characterizing the financial risk resistance of the user;
the determining module is used for determining a first target product matched with the user from a product database according to the user grade; wherein the product database includes a plurality of financial products having tag information.
10. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 8.
11. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 8.
12. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 8 when executed by a processor.
CN202210445601.8A 2022-04-24 2022-04-24 Product matching method, device, computer equipment, storage medium and program product Pending CN114693409A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210445601.8A CN114693409A (en) 2022-04-24 2022-04-24 Product matching method, device, computer equipment, storage medium and program product

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210445601.8A CN114693409A (en) 2022-04-24 2022-04-24 Product matching method, device, computer equipment, storage medium and program product

Publications (1)

Publication Number Publication Date
CN114693409A true CN114693409A (en) 2022-07-01

Family

ID=82144348

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210445601.8A Pending CN114693409A (en) 2022-04-24 2022-04-24 Product matching method, device, computer equipment, storage medium and program product

Country Status (1)

Country Link
CN (1) CN114693409A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994821A (en) * 2023-01-09 2023-04-21 中云融拓数据科技发展(深圳)有限公司 Method for establishing financial wind control system based on industrial chain digital scene financial model

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994821A (en) * 2023-01-09 2023-04-21 中云融拓数据科技发展(深圳)有限公司 Method for establishing financial wind control system based on industrial chain digital scene financial model

Similar Documents

Publication Publication Date Title
CN110837931B (en) Customer churn prediction method, device and storage medium
CN107818344B (en) Method and system for classifying and predicting user behaviors
CN109582876B (en) Tourist industry user portrait construction method and device and computer equipment
CN110110233B (en) Information processing method, device, medium and computing equipment
KR20200039852A (en) Method for analysis of business management system providing machine learning algorithm for predictive modeling
CN113177700B (en) Risk assessment method, system, electronic equipment and storage medium
CN111429161B (en) Feature extraction method, feature extraction device, storage medium and electronic equipment
CN111582932A (en) Inter-scene information pushing method and device, computer equipment and storage medium
CN112633690A (en) Service personnel information distribution method, service personnel information distribution device, computer equipment and storage medium
CN114371946B (en) Information push method and information push server based on cloud computing and big data
CN115080868A (en) Product pushing method, product pushing device, computer equipment, storage medium and program product
CN115983900A (en) Method, apparatus, device, medium, and program product for constructing user marketing strategy
CN115545886A (en) Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium
Liu et al. Extracting, ranking, and evaluating quality features of web services through user review sentiment analysis
CN114997916A (en) Prediction method, system, electronic device and storage medium of potential user
CN114819967A (en) Data processing method and device, electronic equipment and computer readable storage medium
CN114358014A (en) Work order intelligent diagnosis method, device, equipment and medium based on natural language
CN114693409A (en) Product matching method, device, computer equipment, storage medium and program product
CN117312657A (en) Abnormal function positioning method and device for financial application, computer equipment and medium
CN112330442A (en) Modeling method and device based on ultra-long behavior sequence, terminal and storage medium
CN115630221A (en) Terminal application interface display data processing method and device and computer equipment
CN114529399A (en) User data processing method, device, computer equipment and storage medium
CN112328899B (en) Information processing method, information processing apparatus, storage medium, and electronic device
CN112085566B (en) Product recommendation method and device based on intelligent decision and computer equipment
CN114118610A (en) Product sales prediction method and system based on relevance big data

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination