Information recommendation method and device, equipment and medium

Abstract

The invention belongs to the field of artificial intelligence, and relates to an information recommendation method, which comprises the following steps: acquiring customer data and customer behavior feature information of a plurality of customer terminals; constructing client feature vectors of the plurality of client terminals based on the data and the information; a deep clustering model is adopted to perform clustering analysis on the client feature vectors, and a client group to which each client terminal belongs is determined; determining a target sub-customer group from the customer groups based on a marketing strategy, and predicting the probability of adding the instant messaging group to the customer terminal of the target sub-customer group by adopting a prediction model; and adding the target client terminals of which the probability values are greater than a preset probability threshold value in the target sub-client group to the instant messaging group so as to perform information recommendation on the target client terminals in the instant messaging group. The invention further provides an information recommendation device, equipment and a medium. In addition, the invention also relates to a block chain technology, and client data and client behavior feature information can be stored in a block chain. The accuracy of information recommendation can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an information recommendation method, device, equipment and medium. Background Art

[0002] With the rapid development of mobile Internet technology, people's lifestyles and work patterns have undergone profound changes, and information acquisition channels have become increasingly diversified. This change has had a significant impact on the market operation model of traditional service industries such as the insurance industry. Although traditional insurance sales models, such as fixed store services, door-to-door visits and telephone sales, have helped companies accumulate a large user base, their inherent low efficiency and high cost problems have become increasingly prominent in the context of the information age, making it difficult to adapt to changes in market demand.

[0003] In recent years, with the rise of social media platforms, instant messaging tools have become a new channel for companies to expand their markets. Many companies have tried to use these platforms to promote products and introduce services in order to achieve innovation in marketing models. However, this new marketing model is not a panacea. Some customers are troubled by frequent information push and even feel disgusted, resulting in an increase in customer complaints and seriously damaging the company's brand image. In this context, there is an urgent need to provide a method that can accurately identify and distinguish customer groups willing to accept online marketing methods and formulate personalized marketing strategies for them. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose an information recommendation method, device, equipment and medium to solve the problem that it is difficult to accurately identify and distinguish customer groups who are willing to accept online marketing methods, and to formulate personalized marketing strategies for them to recommend information.

[0005] In order to solve the above technical problems, the present application embodiment provides an information recommendation method, which adopts the following technical solutions:

[0006] Acquire customer data and customer behavior characteristic information of multiple customer terminals within a preset time period; construct customer feature vectors of multiple customer terminals based on the customer data and customer behavior characteristic information; use a preset deep clustering model to perform cluster analysis on the customer feature vectors to determine the customer group to which each customer terminal belongs; based on a preset marketing strategy, determine a target sub-customer group from the customer group, use a preset prediction model to predict the probability of adding customer terminals of the target sub-customer group to the instant messaging group, and obtain a probability value; add target customer terminals in the target sub-customer group whose probability values ​​are greater than a preset probability threshold to the instant messaging group, so as to recommend information to the target customer terminals in the instant messaging group.

[0007] Furthermore, the step of obtaining the customer data and customer behavior characteristic information of multiple customer terminals within a preset time period specifically includes:

[0008] Obtain original customer data and customer behavior characteristic information of multiple customer terminals within a preset time period; perform data cleaning on the original customer data to obtain cleaned original customer data, and perform encoding conversion on the cleaned original customer data to obtain customer data of multiple customer terminals.

[0009] Furthermore, the step of constructing customer feature vectors of multiple customer terminals based on the customer data and customer behavior feature information specifically includes:

[0010] A preset dimensionality reduction algorithm is used to perform feature selection on customer data to obtain key feature sets of multiple customer terminals; based on the key feature sets and customer behavior feature information, customer feature vectors of multiple customer terminals are constructed.

[0011] Furthermore, before the step of using a preset deep clustering model to perform cluster analysis on the customer feature vectors to determine the customer group to which each customer terminal belongs, the method further includes:

[0012] Obtain historical customer data and historical behavior feature information of multiple historical customer terminals that have been marked within a first historical time period; construct target feature vectors of multiple historical customer terminals based on the historical customer data and historical behavior feature information; and use the target feature vectors to train a preset multi-layer network and a preset clustering layer to construct a preset deep clustering model.

[0013] Furthermore, the steps of using a preset deep clustering model to perform cluster analysis on the customer feature vectors and determining the customer group to which each customer terminal belongs specifically include:

[0014] A multi-layer network of a deep clustering model is used to perform nonlinear transformation on customer feature vectors to obtain deep feature information of each customer terminal. The clustering layer of the deep clustering model is used to cluster the deep feature information to determine the customer group to which each customer terminal belongs.

[0015] Further, based on the preset marketing strategy, the steps of determining the target sub-customer group from the customer group specifically include:

[0016] According to the preset marketing strategy, obtain the attribute characteristic data of the customer group;

[0017] A clustering algorithm is used to analyze the attribute feature data to obtain the attribute similarity between customer terminals in the customer group; the customer group is divided into multiple sub-customer groups based on the attribute similarity; the network activity of each sub-customer group is obtained, and the sub-customer group whose network activity exceeds a preset activity threshold among the multiple sub-customer groups is marked as the target sub-customer group.

[0018] Furthermore, before the step of using a preset prediction model to predict the probability of a client terminal of a target sub-client group adding an instant messaging group and obtaining a probability value, the method further includes:

[0019] The grouping data of the historical customer group in the second historical time period is obtained; based on the grouping data and a preset supervised learning algorithm, a preset classification model is trained to build a preset prediction model.

[0020] In order to solve the above technical problems, the embodiment of the present application further provides an information recommendation device, which adopts the following technical solution:

[0021] A first acquisition module is used to acquire customer data and customer behavior characteristic information of multiple customer terminals within a preset time period;

[0022] A construction module, used to construct customer feature vectors of multiple customer terminals based on customer data and customer behavior feature information;

[0023] A clustering module is used to perform cluster analysis on customer feature vectors using a preset deep clustering model to determine the customer group to which each customer terminal belongs;

[0024] A prediction module is used to determine a target sub-customer group from the customer group based on a preset marketing strategy, and use a preset prediction model to predict the probability of a customer terminal of the target sub-customer group adding an instant messaging group to obtain a probability value;

[0025] The recommendation module is used to add target client terminals in the target sub-client group whose probability values ​​are greater than a preset probability threshold to the instant messaging group, so as to recommend information to the target client terminals in the instant messaging group.

[0026] In order to solve the above technical problems, an embodiment of the present application further provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the above information recommendation method are implemented.

[0027] In order to solve the above technical problems, an embodiment of the present application also provides a computer-readable storage medium, which stores computer-readable instructions. The computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above information recommendation method.

[0028] Compared with the prior art, the embodiment of the present application has the following beneficial effects: by acquiring and analyzing the customer data and behavior feature information of multiple customer terminals within a preset time period, it is possible to fully and deeply understand customer preferences and needs. By constructing a customer feature vector and using a deep clustering model to accurately divide customer groups, a detailed segmentation of the target market is achieved. Further, based on a preset marketing strategy, the solution provided by this embodiment can lock in a target sub-customer group with specific needs, that is, it can accurately identify and distinguish customer groups willing to accept online marketing methods, and accurately evaluate their willingness to join an instant messaging group through a prediction model. This step not only improves the pertinence of marketing activities, but also effectively avoids unnecessary waste of resources. Finally, the target customer terminal with a high probability value is added to the instant messaging group to recommend information to the target customer in the instant messaging group, so as to formulate a personalized marketing strategy for the target customer terminal with a high probability value, so that the information recommendation can directly reach potential customers, greatly improving the marketing efficiency and customer satisfaction, and improving the accuracy of information recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0031] Figure 2 It is a flowchart of an information recommendation method provided by this application;

[0032] Figure 3 It is a structural schematic diagram of an information recommendation device provided by the present application;

[0033] Figure 4 It is a structural schematic diagram of a computer device provided by this application. DETAILED DESCRIPTION

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0035] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0036] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0037] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0038] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0039] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0040] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0041] It should be noted that the information recommendation method provided in the embodiment of the present application is generally executed by a server, and accordingly, the information recommendation device is generally arranged in the server.

[0042] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0043] Continue to refer Figure 2 , shows a flow chart of an embodiment of an information recommendation method according to the present application. The information recommendation method comprises the following steps:

[0044] Step S201, obtaining customer data and customer behavior characteristic information of multiple customer terminals within a preset time period.

[0045] In this embodiment, the information recommendation method is executed on the electronic device (eg Figure 1 The server shown in the figure) can obtain the customer data and customer behavior characteristic information of multiple client terminals within a preset time period through a wired connection or a wireless connection. It should be noted that the above wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0046] The preset time period refers to a specific time range set in advance and used to collect and analyze data. It is used to ensure the timeliness and relevance of customer data and customer behavior feature information, and facilitate subsequent customer behavior analysis and feature extraction.

[0047] The multiple client terminals refer to various electronic devices used by different users, including but not limited to smart phones, tablet computers, personal computers, etc. These devices are used to access services or applications and generate customer data.

[0048] Among them, customer data refers to basic information about customers (such as age, gender, occupation, etc.), behavioral data (such as historical transaction records, interaction frequency, interaction methods, APP activity, etc.) and other relevant third-party data collected from multiple customer terminals, and obtained after pre-processing of these data.

[0049] Among them, customer behavior characteristic information describes the behavior patterns exhibited by customers when using services or applications, such as visit frequency, stay time, purchase preferences, behavioral habits, etc., which is used to gain a deeper understanding of customer needs.

[0050] Step S202: constructing customer feature vectors of multiple customer terminals based on customer data and customer behavior feature information.

[0051] Among them, the customer feature vector refers to the vector form in which customer data and customer behavior feature information are converted through mathematical methods. It can be used to represent the comprehensive characteristics of customers, which is convenient for subsequent customer group division and cluster analysis.

[0052] Step S203: Use a preset deep clustering model to perform cluster analysis on the customer feature vectors to determine the customer group to which each customer terminal belongs.

[0053] Among them, the deep clustering model is a clustering model built based on deep learning technology, which can automatically learn the deep-level characteristics of the data and achieve more accurate customer group segmentation.

[0054] Cluster analysis is an unsupervised learning method used to group similar data points into one category, thereby discovering potential structures and patterns in the data. For example, customer feature vectors are grouped to determine the customer group to which each customer terminal belongs.

[0055] Among them, customer group refers to a collection of customers with similar characteristics or behavior patterns, which is obtained through cluster analysis and facilitates the formulation of targeted information recommendation methods.

[0056] Step S204, based on the preset marketing strategy, determine the target sub-customer group from the customer group, use the preset prediction model to predict the probability of the client terminal of the target sub-customer group adding the instant messaging group, and obtain the probability value.

[0057] Among them, marketing strategy refers to the plans and methods formulated by enterprises to achieve marketing goals. It can be used to guide the segmentation of customer groups and the implementation of marketing activities.

[0058] Among them, the target sub-customer group refers to the sub-group with network marketing value selected based on the customer group according to specific marketing strategies, and is the focus of marketing activities.

[0059] The prediction model is a model trained based on historical data, which can be used to predict the probability or result of future events, that is, to predict the probability of adding instant messaging groups to customer terminals of the target sub-customer group.

[0060] Among them, instant messaging groups refer to an online communication platform that allows multiple users to communicate in real time through text, voice, video, etc. It can be used to push information to customer terminals of target sub-customer groups and conduct marketing interactions.

[0061] The probability value refers to the predicted probability value output by the prediction model that the client terminal of the target sub-client group will add an instant messaging group.

[0062] Step S205 , adding the target client terminals in the target sub-client group whose probability values ​​are greater than a preset probability threshold to the instant messaging group, so as to recommend information to the target client terminals in the instant messaging group.

[0063] The preset probability threshold is used to determine whether a client terminal in the target sub-client group is likely to join the instant messaging group. When the predicted probability value exceeds this threshold, it is considered that the client terminal has a high probability of joining the group.

[0064] Among them, the target customer terminal refers to the electronic device used by customers who have been screened and predicted and are considered likely to join the instant messaging group and interact with information.

[0065] Among them, information recommendation refers to the process of recommending relevant product or service information to target customer terminals within an instant messaging group based on their historical behaviors and preferences, with the aim of improving customer satisfaction and marketing effectiveness.

[0066] The embodiment of the present application can fully and deeply understand customer preferences and needs by acquiring and analyzing customer data and behavioral feature information of multiple client terminals within a preset time period. By constructing a customer feature vector and using a deep clustering model to accurately divide customer groups, a detailed segmentation of the target market is achieved. Further, based on a preset marketing strategy, the solution provided by this embodiment can lock in a target sub-customer group with specific needs, that is, it can accurately identify and distinguish customer groups willing to accept online marketing methods, and accurately evaluate their willingness to join the instant messaging group through a prediction model. This step not only improves the pertinence of marketing activities, but also effectively avoids unnecessary waste of resources. Finally, the target customer terminal with a high probability value is added to the instant messaging group to recommend information to the target customer in the instant messaging group, so as to formulate a personalized marketing strategy for the target customer terminal with a high probability value, so that the information recommendation can directly reach potential customers, greatly improving marketing efficiency and customer satisfaction, and improving the accuracy of information recommendation.

[0067] In some optional implementations of this embodiment, step 201, obtaining customer data and customer behavior characteristic information of multiple customer terminals within a preset time period, specifically includes the following steps:

[0068] Obtain original customer data and customer behavior characteristic information of multiple customer terminals within a preset time period; perform data cleaning on the original customer data to obtain cleaned original customer data, and perform encoding conversion on the cleaned original customer data to obtain customer data of multiple customer terminals.

[0069] Among them, original customer data refers to unprocessed customer data collected directly from customer terminals, including various original information.

[0070] The cleaned original customer data refers to the customer data that has been cleaned and outliers and missing values ​​have been removed, which can improve data quality and analysis accuracy.

[0071] Among them, encoding conversion refers to converting the cleaned original customer data into a format suitable for subsequent analysis, which is convenient for data processing and feature extraction.

[0072] In one example, basic information (such as age, gender, occupation, etc.), behavioral data (such as historical transaction records, interaction frequency, interaction method, APP activity, etc.) and other relevant third-party data corresponding to multiple customer terminals within a preset time period can be extracted from the insurance company's customer relationship management system. The collected data is cleaned to remove outliers and missing values, and non-numeric data is encoded and converted to facilitate subsequent algorithm processing, thereby obtaining customer data of multiple customer terminals.

[0073] In one example, the original customer data of multiple customer terminals (such as individual users' mobile banking apps, online banking accounts, etc.) within a preset time period (such as the past year) can be obtained through the built-in data collection module of the bank system, including but not limited to transaction records, account balances, loan application history, etc. At the same time, the system also collects customer behavior feature information, such as login frequency, transaction time preference, browsing history, etc., to fully reflect the customer's financial activity habits. Then, using machine learning algorithms, identify and eliminate abnormal transaction records (such as large non-daily consumption), duplicate records and invalid information to ensure the accuracy and consistency of the data. For missing values, mean interpolation or prediction based on similar customer behavior patterns is used. The cleaned original customer data is then encoded and converted, such as converting the transaction type into a standardized code (such as "P001" for housing loan repayment), and the customer identity information is processed through hash encryption to protect personal privacy while facilitating subsequent analysis, thereby obtaining customer data of multiple customer terminals.

[0074] The embodiment of the present application can ensure the comprehensiveness and timeliness of the data by fully acquiring the original customer data and customer behavior characteristic information of multiple customer terminals within a preset time period, providing a solid foundation for subsequent data analysis. Secondly, the original customer data is cleaned to effectively eliminate invalid, redundant and abnormal data, improve the purity and availability of the data, and reduce interference for subsequent data analysis. Finally, the cleaned original customer data is encoded and converted, which not only realizes the standardization and normalization of the data, but also protects the privacy and security of the customer, while facilitating the storage, transmission and processing of the data.

[0075] In some optional implementations of this embodiment, in step S202, customer feature vectors of multiple customer terminals are constructed based on customer data and customer behavior feature information, specifically including the following steps:

[0076] A preset dimensionality reduction algorithm is used to perform feature selection on customer data to obtain key feature sets of multiple customer terminals; based on the key feature sets and customer behavior feature information, customer feature vectors of multiple customer terminals are constructed.

[0077] Among them, the dimensionality reduction algorithm is a technology used to reduce the number of data features. By retaining the main information of the data, it reduces data complexity and computational costs and improves analysis efficiency.

[0078] Feature selection refers to the process of selecting features that have a significant impact on the target variable from customer data, which is used to construct a more concise and effective customer feature vector.

[0079] Among them, the key feature set refers to the feature set obtained through feature selection that has an important impact on customer group segmentation and clustering analysis.

[0080] In one example, first, customer data of multiple customer terminals that have undergone preliminary processing (such as data cleaning and encoding conversion) are received. Subsequently, a preset dimensionality reduction algorithm, such as principal component analysis (PCA), linear discriminant analysis (LDA) or random forest feature selection, is used to perform feature selection on these customer data to identify and extract the most representative key feature set. On this basis, combined with customer behavior feature information, such as browsing habits, purchase preferences, interaction frequency, behavioral habits, etc., a customer feature vector for each customer terminal is jointly constructed.

[0081] The embodiment of the present application performs feature selection on customer data through a preset dimensionality reduction algorithm, which effectively reduces the dimension of customer data and reduces redundant information, thereby improving the efficiency of customer data processing. At the same time, the dimensionality reduction algorithm can identify and extract the most representative key feature sets, which can more accurately reflect the essential attributes and behavior patterns of customers. Secondly, based on the key feature set and customer behavior feature information, a customer feature vector is constructed, which not only integrates the static attributes and dynamic behavior characteristics of the customer, but also realizes a comprehensive and accurate description of the customer's characteristics.

[0082] In some optional implementations of this embodiment, in step S203, before performing cluster analysis on the customer feature vectors using a preset deep clustering model to determine the customer group to which each customer terminal belongs, the following steps are specifically included:

[0083] Obtain historical customer data and historical behavior feature information of multiple historical customer terminals that have been marked within a first historical time period; construct target feature vectors of multiple historical customer terminals based on the historical customer data and historical behavior feature information; and use the target feature vectors to train a preset multi-layer network and a preset clustering layer to construct a preset deep clustering model.

[0084] The first historical time period refers to a specific time range for collecting the marked historical data of the target client terminal.

[0085] The historical customer data refers to the customer data of multiple historical customer terminals collected in a certain period of time in the past, which is used to construct the target feature vector and train the prediction model.

[0086] The historical behavior characteristic information refers to the behavior patterns exhibited by customers of multiple historical customer terminals in the process of using services or applications in the first historical time period, such as access frequency, stay time, purchase preferences, etc.

[0087] Among them, the target feature vector refers to the customer feature vector constructed based on historical customer data and historical behavior feature information, which is used to train the deep clustering model.

[0088] In one example, through deep learning methods, the labeled data (such as historical customer data and historical behavior feature information) in the historical time period can be deeply mined to build an accurate customer segmentation model, that is, a deep clustering model. Specifically, first, the historical customer data and historical behavior feature information of multiple historical customer terminals that have been labeled in the first historical time period (such as the past year) can be obtained. Then, based on these data, a target feature vector is constructed for each historical customer terminal, and the target feature vector comprehensively reflects the static attributes and dynamic behavior characteristics of the customer. Finally, using these target feature vectors, the preset multi-layer network and clustering layer are trained to build a deep clustering model that can automatically perform customer segmentation.

[0089] The embodiment of the present application can construct a target feature vector by utilizing the data of multiple historical customer terminals that have been marked in the first historical time period. The target feature vector comprehensively reflects the static attributes and dynamic behavior characteristics of the customer, providing a rich information basis for subsequent customer segmentation. Secondly, the target feature vector is used to train the preset multi-layer network and clustering layer to construct a deep clustering model. The deep clustering model can automatically learn the potential characteristics and distribution patterns of customers and realize the accurate division of customer groups. Compared with traditional clustering algorithms, this deep clustering method has higher clustering accuracy and robustness, and can better adapt to complex and changing customer data environments.

[0090] In some optional implementations of this embodiment, step S203, using a preset deep clustering model to perform cluster analysis on the customer feature vectors to determine the customer group to which each customer terminal belongs, specifically includes the following steps:

[0091] A multi-layer network of a deep clustering model is used to perform nonlinear transformation on customer feature vectors to obtain deep feature information of each customer terminal. The clustering layer of the deep clustering model is used to cluster the deep feature information to determine the customer group to which each customer terminal belongs.

[0092] Among them, nonlinear transformation refers to a complex transformation of the customer feature vector by the multi-layer network of the deep clustering model, which can capture the nonlinear relationship in the customer feature vector.

[0093] Among them, deep feature information refers to the customer feature information obtained after nonlinear transformation of the multi-layer network of the deep clustering model, which contains the deep structure and pattern of the customer feature vector.

[0094] In one example, the multi-layer network of the deep clustering model can be used to perform nonlinear transformation on the customer feature vector. The multi-layer network consists of multiple hidden layers, each hidden layer contains multiple neurons, and nonlinear mapping is achieved through activation functions. Through layer-by-layer abstraction and feature extraction, the deep feature information of the customer is learned, which can more accurately reflect the customer's potential needs and preferences. After obtaining the deep feature information, the clustering layer of the deep clustering model is used for clustering operations. The clustering layer can select clustering algorithms such as K-means and DBSCAN to divide the customer terminals into different customer groups according to the similarity of the deep feature information. Output the customer group information to which each customer terminal belongs, and determine the customer group to which each customer terminal belongs.

[0095] The multi-layer network of the deep clustering model provided in the embodiment of the present application can make full use of its powerful nonlinear transformation ability, deeply explore the potential information in the customer feature vector, and thus extract the deep feature information of each customer terminal. Compared with the original customer feature vector, these deep feature information can more accurately reflect the customer's intrinsic attributes and demand preferences. Subsequently, the clustering layer of the deep clustering model performs clustering based on these deep feature information, which can more accurately divide customer groups. This customer segmentation method based on deep learning not only improves the accuracy of customer group division, but also enhances the robustness and adaptability of the model, making the customer group division results more stable and reliable.

[0096] In some optional implementations of this embodiment, step S204, based on a preset marketing strategy, determining a target sub-customer group from a customer group, may specifically include the following steps:

[0097] According to the preset marketing strategy, the attribute feature data of the customer group is obtained; the attribute feature data is analyzed by using a clustering algorithm to obtain the attribute similarity between customer terminals in the customer group; the customer group is divided into multiple sub-customer groups according to the attribute similarity; the network activity of each sub-customer group is obtained, and the sub-customer group whose network activity exceeds the preset activity threshold among the multiple sub-customer groups is marked as the target sub-customer group.

[0098] Among them, attribute feature data refers to data that describes the attributes of customer groups, such as age attributes, gender attributes, income levels, consumption habits, etc. It is used to further segment customer groups.

[0099] Among them, attribute similarity refers to an indicator that measures the similarity between attribute characteristics of each customer terminal in a customer group, which can be used to divide customer groups.

[0100] Among them, multiple sub-customer groups refer to smaller customer groups obtained by further segmenting the customer group based on attribute characteristic data.

[0101] Among them, network activity refers to an indicator that measures the level of customer activity on the online platform, such as visit frequency, number of interactions, etc., which is used to evaluate the marketing potential of sub-customer groups.

[0102] Among them, the preset active threshold is used to determine whether the network activity of the sub-customer group has reached a relatively high standard value, and is used to screen the target sub-customer group.

[0103] In one example, first, the attribute feature data of the customer group can be extracted from the database according to the preset marketing strategy. These attribute feature data are preprocessed and cleaned to ensure the accuracy and consistency of the data. Next, the attribute feature data can be analyzed by a clustering algorithm. In the clustering process, the attribute similarity between customer terminals, such as Euclidean distance, cosine similarity, etc., is calculated, and the customer group is divided into multiple sub-customer groups according to the similarity. The clustering results are displayed through a visualization tool so that the distribution of the customer group can be intuitively understood. Next, the network activity of each sub-customer group can be calculated by monitoring the customer's behavior on the network platform (such as browsing, clicking, purchasing, etc.). Network activity can be measured by statistically analyzing indicators such as the frequency and duration of customer behavior. Then, the network activity of each sub-customer group is compared with the preset active threshold, and the sub-customer group whose network activity exceeds the threshold is marked as the target sub-customer group.

[0104] The embodiments of the present application can accurately obtain and analyze the attribute characteristic data of the customer group according to the preset marketing strategy. The effective use of the clustering algorithm enables the accurate calculation of the attribute similarity between customer terminals, thereby realizing the refined division of the customer group and forming multiple sub-customer groups with distinct characteristics. On this basis, by monitoring and calculating the network activity of each sub-customer group, the target sub-customer group with high network activity and high potential value can be quickly identified. In this process, the setting of the preset active threshold can ensure the accuracy and effectiveness of the screening of the target sub-customer group.

[0105] In some optional implementations of this embodiment, in step S204, a preset prediction model is used to predict the probability of a client terminal of a target sub-client group adding an instant messaging group, and before obtaining the probability value, the following steps are specifically included:

[0106] The grouping data of the historical customer group in the second historical time period is obtained; based on the grouping data and a preset supervised learning algorithm, a preset classification model is trained to build a preset prediction model.

[0107] The second historical time period refers to a specific time range for collecting historical customer group and group data for training the prediction model.

[0108] The historical customer group refers to the customer group that already existed in the second historical time period, and its group data is used to train the prediction model.

[0109] The group joining data refers to the data describing the situation of customer members of a customer group joining an instant messaging group, which can be used to train a classification model.

[0110] Among them, supervised learning algorithm is a machine learning algorithm. The classification model is trained through known output results (labels) so that it can predict the output results of new data after training.

[0111] In one example, first, the grouping data of the historical customer group in the second historical time period is extracted from the database, and the grouping data may include basic information of the customer group (such as age, gender, region, etc.), grouping time, grouping type, etc. Then, based on the grouping data and the preset supervised learning algorithm (such as logistic regression, support vector machine, random forest, etc.), the preset classification model is trained to build a prediction model. During the training process, the classification model parameters are continuously adjusted to optimize the classification model performance until the preset accuracy or number of iterations is reached to obtain a prediction model.

[0112] The historical customer group joining data in the second historical time period obtained by the embodiment of the present application covers the customer group joining behavior characteristics, and provides a rich sample for the subsequent classification model training. Based on these data, by applying a preset supervised learning algorithm, a preset classification model is trained to construct an accurate prediction model. The prediction model can accurately predict future group joining behavior based on the historical group joining behavior characteristics of the customer group.

[0113] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned customer data and customer behavior characteristic information, the above-mentioned customer data and customer behavior characteristic information can also be stored in a node of a blockchain.

[0114] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0115] The embodiments of the present application can build and optimize related models and networks based on artificial intelligence technology, such as deep clustering models, prediction models, etc. Among them, artificial intelligence (AI) models are the crystallization of theory and practice that simulates the decision-making process of human intelligence through algorithms and data analysis to solve complex problems, predict future trends, or realize automation tasks. These models use a large amount of historical data and real-time information, and are trained and optimized through a specific algorithm framework to achieve efficient, accurate and reliable performance.

[0116] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0117] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0118] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of an information recommendation device, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0119] like Figure 3 As shown, the information recommendation device 400 of this embodiment includes: a first acquisition module 401 , a construction module 402 , a clustering module 403 , a prediction module 404 and a recommendation module 405 .

[0120] in:

[0121] The first acquisition module 401 is used to acquire customer data and customer behavior characteristic information of multiple customer terminals within a preset time period;

[0122] A construction module 402 is used to construct customer feature vectors of multiple customer terminals based on customer data and customer behavior feature information;

[0123] The clustering module 403 is used to perform cluster analysis on the customer feature vectors using a preset deep clustering model to determine the customer group to which each customer terminal belongs;

[0124] Prediction module 404, used to determine a target sub-customer group from the customer group based on a preset marketing strategy, and use a preset prediction model to predict the probability of a customer terminal of the target sub-customer group adding an instant messaging group to obtain a probability value;

[0125] The recommendation module 405 is used to add the target client terminals in the target sub-client group whose probability values ​​are greater than a preset probability threshold to the instant messaging group, so as to recommend information to the target client terminals in the instant messaging group.

[0126] The embodiment of the present application can fully and deeply understand customer preferences and needs by acquiring and analyzing customer data and behavioral feature information of multiple client terminals within a preset time period. By constructing a customer feature vector and using a deep clustering model to accurately divide customer groups, a detailed segmentation of the target market is achieved. Further, based on a preset marketing strategy, the solution provided by this embodiment can lock in a target sub-customer group with specific needs, that is, it can accurately identify and distinguish customer groups willing to accept online marketing methods, and accurately evaluate their willingness to join the instant messaging group through a prediction model. This step not only improves the pertinence of marketing activities, but also effectively avoids unnecessary waste of resources. Finally, the target customer terminal with a high probability value is added to the instant messaging group to recommend information to the target customer in the instant messaging group, so as to formulate a personalized marketing strategy for the target customer terminal with a high probability value, so that the information recommendation can directly reach potential customers, greatly improving marketing efficiency and customer satisfaction, and improving the accuracy of information recommendation.

[0127] In one embodiment, the first acquisition module 401 includes:

[0128] The first acquisition submodule is used to acquire original customer data and customer behavior characteristic information of multiple customer terminals within a preset time period;

[0129] The cleaning submodule is used to perform data cleaning processing on the original customer data to obtain the cleaned original customer data, and perform code conversion on the cleaned original customer data to obtain customer data of multiple customer terminals.

[0130] The embodiment of the present application can ensure the comprehensiveness and timeliness of the data by fully acquiring the original customer data and customer behavior characteristic information of multiple customer terminals within a preset time period, providing a solid foundation for subsequent data analysis. Secondly, the original customer data is cleaned to effectively eliminate invalid, redundant and abnormal data, improve the purity and availability of the data, and reduce interference for subsequent data analysis. Finally, the cleaned original customer data is encoded and converted, which not only realizes the standardization and normalization of the data, but also protects the privacy and security of the customer, while facilitating the storage, transmission and processing of the data.

[0131] In one embodiment, the building block 402 includes:

[0132] The feature selection submodule is used to select features of customer data using a preset dimensionality reduction algorithm to obtain key feature sets of multiple customer terminals;

[0133] The construction submodule is used to construct customer feature vectors of multiple customer terminals based on the key feature set and customer behavior feature information.

[0134] The embodiment of the present application can perform feature selection on customer data through a preset dimensionality reduction algorithm, effectively reducing the dimension of customer data and reducing redundant information, thereby improving the efficiency of customer data processing. At the same time, the dimensionality reduction algorithm can identify and extract the most representative key feature sets, which can more accurately reflect the essential attributes and behavior patterns of customers. Secondly, based on the key feature set and customer behavior feature information, a customer feature vector is constructed, which not only integrates the static attributes and dynamic behavior characteristics of the customer, but also realizes a comprehensive and accurate description of the customer's characteristics.

[0135] In one embodiment, the information recommendation device may further include:

[0136] A second acquisition module is used to acquire historical customer data and historical behavior feature information of a plurality of historical customer terminals marked in a first historical time period;

[0137] A vector construction module, used to construct target feature vectors of multiple historical customer terminals based on historical customer data and historical behavior feature information;

[0138] The first training module is used to train a preset multi-layer network and a preset clustering layer using a target feature vector to construct a preset deep clustering model.

[0139] The embodiment of the present application constructs a target feature vector by utilizing the data of multiple historical customer terminals that have been marked in the first historical time period. The target feature vector comprehensively reflects the static attributes and dynamic behavior characteristics of the customer, providing a rich information basis for subsequent customer segmentation. Secondly, the target feature vector is used to train the preset multi-layer network and clustering layer to construct a deep clustering model. The deep clustering model can automatically learn the potential characteristics and distribution patterns of customers and realize the accurate division of customer groups. Compared with traditional clustering algorithms, this deep clustering method has higher clustering accuracy and robustness, and can better adapt to complex and changing customer data environments.

[0140] In one embodiment, the clustering module 403 includes:

[0141] The linear transformation submodule is used to perform nonlinear transformation on the customer feature vector using a multi-layer network of a deep clustering model to obtain deep feature information of each customer terminal;

[0142] The clustering submodule is used to cluster the deep feature information using the clustering layer of the deep clustering model to determine the customer group to which each customer terminal belongs.

[0143] The multi-layer network of the deep clustering model provided in the embodiment of the present application can make full use of its powerful nonlinear transformation ability, deeply explore the potential information in the customer feature vector, and thus extract the deep feature information of each customer terminal. Compared with the original customer feature vector, these deep feature information can more accurately reflect the customer's intrinsic attributes and demand preferences. Subsequently, the clustering layer of the deep clustering model performs clustering based on these deep feature information, which can more accurately divide customer groups. This customer segmentation method based on deep learning not only improves the accuracy of customer group division, but also enhances the robustness and adaptability of the model, making the customer group division results more stable and reliable.

[0144] In one embodiment, the prediction module 404 includes:

[0145] The second acquisition submodule is used for the feature extraction submodule, which is used to obtain the attribute feature data of the customer group according to the preset marketing strategy;

[0146] The analysis submodule is used to analyze the attribute feature data using a clustering algorithm to obtain the attribute similarity between the customer terminals in the customer group;

[0147] A division sub-module is used to divide the customer group into multiple sub-customer groups according to attribute similarity;

[0148] The marking submodule is used to obtain the network activity of each sub-customer group, and mark the sub-customer group whose network activity exceeds a preset activity threshold among multiple sub-customer groups as the target sub-customer group.

[0149] The embodiments of the present application can accurately obtain and analyze the attribute characteristic data of the customer group according to the preset marketing strategy. The effective use of the clustering algorithm enables the accurate calculation of the attribute similarity between customer terminals, thereby realizing the refined division of the customer group and forming multiple sub-customer groups with distinct characteristics. On this basis, by monitoring and calculating the network activity of each sub-customer group, the target sub-customer group with high network activity and high potential value can be quickly identified. In this process, the setting of the preset active threshold can ensure the accuracy and effectiveness of the screening of the target sub-customer group.

[0150] In one embodiment, the information recommendation device may further include:

[0151] A third acquisition module is used to obtain group adding data of the historical customer group in the second historical time period;

[0152] The second training module is used to train the preset classification model based on the group data and the preset supervised learning algorithm to build a preset prediction model.

[0153] The historical customer group joining data in the second historical time period obtained by the embodiment of the present application covers the customer group joining behavior characteristics, and provides a rich sample for the subsequent classification model training. Based on these data, by applying a preset supervised learning algorithm, a preset classification model is trained to construct an accurate prediction model. The prediction model can accurately predict future group joining behavior based on the historical group joining behavior characteristics of the customer group.

[0154] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0155] The computer device 6 includes a memory 61, a processor 62, and a network interface 63 that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 6 having a memory 61, a processor 62, and a network interface 63, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASIC), programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.

[0156] Computer devices can be computing devices such as desktop computers, notebooks, PDAs, and cloud servers. Computer devices can interact with users through keyboards, mice, remote controls, touch pads, or voice control devices.

[0157] The memory 61 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 61 can be an internal storage unit of the computer device 6, such as a hard disk or memory of the computer device 6. In other embodiments, the memory 61 can also be an external storage device of the computer device 6, such as a plug-in hard disk equipped on the computer device 6, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 61 can also include both the internal storage unit of the computer device 6 and its external storage device. In this embodiment, the memory 61 is generally used to store the operating system and various application software installed on the computer device 6, such as computer-readable instructions of the information recommendation method, etc. In addition, the memory 61 can also be used to temporarily store various types of data that have been output or are to be output.

[0158] The processor 62 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 62 is generally used to control the overall operation of the computer device 6. In this embodiment, the processor 62 is used to run computer-readable instructions stored in the memory 61 or process data, such as computer-readable instructions for running the information recommendation method.

[0159] The network interface 63 may include a wireless network interface or a wired network interface, and the network interface 63 is generally used to establish a communication connection between the computer device 6 and other electronic devices.

[0160] The embodiment of the present application can fully and deeply understand customer preferences and needs by acquiring and analyzing customer data and behavioral feature information of multiple client terminals within a preset time period. By constructing a customer feature vector and using a deep clustering model to accurately divide customer groups, a detailed segmentation of the target market is achieved. Further, based on a preset marketing strategy, the solution provided by this embodiment can lock in a target sub-customer group with specific needs, that is, it can accurately identify and distinguish customer groups willing to accept online marketing methods, and accurately evaluate their willingness to join the instant messaging group through a prediction model. This step not only improves the pertinence of marketing activities, but also effectively avoids unnecessary waste of resources. Finally, the target customer terminal with a high probability value is added to the instant messaging group to recommend information to the target customer in the instant messaging group, so as to formulate a personalized marketing strategy for the target customer terminal with a high probability value, so that the information recommendation can directly reach potential customers, greatly improving marketing efficiency and customer satisfaction, and improving the accuracy of information recommendation.

[0161] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores computer-readable instructions. The computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the information recommendation method as described above.

[0162] The embodiment of the present application can fully and deeply understand customer preferences and needs by acquiring and analyzing customer data and behavioral feature information of multiple client terminals within a preset time period. By constructing a customer feature vector and using a deep clustering model to accurately divide customer groups, a detailed segmentation of the target market is achieved. Further, based on a preset marketing strategy, the solution provided by this embodiment can lock in a target sub-customer group with specific needs, that is, it can accurately identify and distinguish customer groups willing to accept online marketing methods, and accurately evaluate their willingness to join the instant messaging group through a prediction model. This step not only improves the pertinence of marketing activities, but also effectively avoids unnecessary waste of resources. Finally, the target customer terminal with a high probability value is added to the instant messaging group to recommend information to the target customer in the instant messaging group, so as to formulate a personalized marketing strategy for the target customer terminal with a high probability value, so that the information recommendation can directly reach potential customers, greatly improving marketing efficiency and customer satisfaction, and improving the accuracy of information recommendation.

[0163] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0164] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. An information recommendation method, characterized in that: The steps include: Obtain customer data and customer behavior characteristic information of multiple customer terminals within a preset time period; constructing customer feature vectors of the plurality of customer terminals based on the customer data and the customer behavior feature information; Using a preset deep clustering model, cluster analysis is performed on the customer feature vectors to determine the customer group to which each customer terminal belongs; Based on a preset marketing strategy, a target sub-customer group is determined from the customer group, and a preset prediction model is used to predict the probability of a customer terminal of the target sub-customer group adding an instant messaging group to obtain a probability value; The target client terminals in the target sub-client group whose probability values ​​are greater than a preset probability threshold are added to the instant messaging group, so as to recommend information to the target client terminals in the instant messaging group.

2. The method according to claim 1, characterized in that The step of obtaining the customer data and customer behavior characteristic information of multiple customer terminals within a preset time period specifically includes: Obtaining original customer data and customer behavior characteristic information of multiple customer terminals within a preset time period; The original customer data is cleansed to obtain cleaned original customer data, and the cleaned original customer data is converted into codes to obtain customer data of the plurality of customer terminals.

3. The method according to any one of claims 1 to 2, characterized in that: The step of constructing the customer feature vectors of the plurality of customer terminals based on the customer data and the customer behavior feature information specifically includes: Using a preset dimensionality reduction algorithm to perform feature selection on the customer data to obtain a key feature set of the multiple customer terminals; Based on the key feature set and the customer behavior feature information, customer feature vectors of the multiple client terminals are constructed.

4. The method according to claim 1, characterized in that: Before the step of using a preset deep clustering model to perform cluster analysis on the customer feature vectors to determine the customer group to which each customer terminal belongs, the method further includes: Acquire historical customer data and historical behavior characteristic information of a plurality of historical customer terminals marked in a first historical time period; Based on the historical customer data and historical behavior feature information, construct target feature vectors of the multiple historical customer terminals; The target feature vector is used to train a preset multi-layer network and a preset clustering layer to construct a preset deep clustering model.

5. The method according to claim 4, characterized in that The step of using a preset deep clustering model to perform cluster analysis on the customer feature vectors to determine the customer group to which each customer terminal belongs specifically includes: Using the multi-layer network of the deep clustering model, non-linearly transform the customer feature vector to obtain deep feature information of each customer terminal; The clustering layer of the deep clustering model is used to cluster the deep feature information to determine the customer group to which each client terminal belongs.

6. The method according to claim 1, characterized in that The step of determining a target sub-customer group from the customer group based on a preset marketing strategy specifically includes: According to a preset marketing strategy, obtaining attribute characteristic data of the customer group; Using a clustering algorithm to analyze the attribute feature data to obtain the attribute similarity between the customer terminals in the customer group; According to the attribute similarity, the customer group is divided into a plurality of sub-customer groups; The network activity of each sub-customer group is obtained, and a sub-customer group whose network activity exceeds a preset activity threshold among the multiple sub-customer groups is marked as a target sub-customer group.

7. The method according to claim 1, characterized in that Before the step of using a preset prediction model to predict the probability of a client terminal of the target sub-client group adding an instant messaging group and obtaining a probability value, the method further includes: Obtain group-added data of the historical customer group in the second historical time period; Based on the grouping data and a preset supervised learning algorithm, a preset classification model is trained to construct a preset prediction model.

8. An information recommendation device, characterized in that: include: A first acquisition module is used to acquire customer data and customer behavior characteristic information of multiple customer terminals within a preset time period; A construction module, configured to construct customer feature vectors of the plurality of customer terminals based on the customer data and the customer behavior feature information; A clustering module, used to perform cluster analysis on the customer feature vectors using a preset deep clustering model to determine the customer group to which each customer terminal belongs; A prediction module, for determining a target sub-customer group from the customer group based on a preset marketing strategy, and using a preset prediction model to predict the probability of a customer terminal of the target sub-customer group adding an instant messaging group to obtain a probability value; The recommendation module is used to add the target client terminals in the target sub-client group whose probability values ​​are greater than a preset probability threshold to the instant messaging group, so as to recommend information to the target client terminals in the instant messaging group.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the information recommendation method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the information recommendation method according to any one of claims 1 to 7 are implemented.

Images