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 steps of determining a breakpoint user terminal based on acquired user active data, and acquiring breakpoint user information of the breakpoint user terminal; obtaining a target user portrait based on the user portrait and the breakpoint user information; based on the target user portrait, performing behavior prediction on the breakpoint user terminal by adopting an evaluation model to obtain a behavior prediction value; calculating the matching degree between the processing terminal and the breakpoint user terminal based on the historical sales data and the target user portrait; and recommending user information to the processing terminal based on the matching degree and the behavior prediction value. The invention further provides an information recommendation device, equipment and a medium. In addition, the invention also relates to a block chain technology, and the user active data, the breakpoint user information, the user portrait, the target user portrait, the behavior prediction value and the historical sales data can be stored in a block chain. The accuracy of user 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] Driven by the wave of mobile Internet, the insurance industry is undergoing an unprecedented digital transformation. The rise of online platforms such as insurance apps and mini-programs has not only reshaped the delivery model of insurance services, but also greatly broadened the market reach of insurance products, allowing consumers to obtain insurance consultation, purchase and claims services anytime and anywhere, greatly improving user experience and satisfaction.

[0003] However, with the explosive growth of online user groups, how to efficiently and accurately match customer needs with insurance products and services has become a major challenge facing insurance companies. Currently, the mechanism for recommending customer information to insurance agents is often limited by problems such as single data dimensions and insufficiently refined analysis models, resulting in low recommendation accuracy. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide an information recommendation method, device, equipment and medium to solve the problem of low accuracy of existing user information recommendation.

[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] Obtain user activity data of the user terminal in the target application; based on the user activity data, determine the breakpoint user terminal from the user terminals, and obtain the breakpoint user information of the breakpoint user terminal; based on the preset user portrait and breakpoint user information, obtain the target user portrait corresponding to the breakpoint user terminal; based on the target user portrait, use the preset evaluation model to predict the behavior of the breakpoint user terminal to obtain the behavior prediction value; obtain the historical sales data of the processing terminal, and based on the historical sales data and the target user portrait, calculate the matching degree between the processing terminal and the breakpoint user terminal; based on the matching degree and the behavior prediction value, recommend user information to the processing terminal.

[0007] Further, based on the user activity data, the step of determining the breakpoint user terminal from the user terminals specifically includes:

[0008] The user activity data is parsed to obtain the breakpoint data, user identification and behavior data of the user terminal; based on the breakpoint data, user identification and behavior data, a user status table of the user terminal is constructed; a preset behavior threshold is compared with the behavior data in the user status table to obtain a comparison result; based on the comparison result, it is determined whether the activity level of the user terminal reaches the behavior threshold; if the activity level does not reach the behavior threshold, the user terminal is marked as a breakpoint user terminal.

[0009] Furthermore, before the step of acquiring the target user portrait corresponding to the breakpoint user terminal based on the preset user portrait and the breakpoint user information, the method further includes:

[0010] Obtain historical user data from the target application, preprocess the historical user data to obtain a processed user data set; extract features from the data in the user data set to obtain user feature information; and build a preset user portrait based on the user feature information.

[0011] Furthermore, the step of constructing a preset user portrait based on the user characteristic information specifically includes:

[0012] Match the user feature information with the preset user portrait template; if the user feature information matches the user portrait template, determine the user portrait template as the preset user portrait; if the user feature information does not match the user portrait template, use the portrait construction algorithm to construct the portrait based on the user feature information to generate the preset user portrait.

[0013] Further, based on the preset user portrait and breakpoint user information, the step of obtaining the target user portrait corresponding to the breakpoint user terminal specifically includes:

[0014] Perform behavioral feature extraction on the breakpoint user information to obtain the behavioral feature information of the breakpoint user terminal; match the behavioral feature information with the feature label information of the user portrait to obtain a matching result; based on the matching result, determine the candidate user portrait corresponding to the breakpoint user terminal from the user portrait; associate the candidate user portrait with the behavioral feature information to generate a target user portrait of the breakpoint user terminal.

[0015] Furthermore, based on the historical sales data and the target user portrait, the step of calculating the matching degree between the processing terminal and the breakpoint user terminal specifically includes:

[0016] Extract product attribute feature information of historical sales data; calculate the similarity between the product attribute feature information and the feature label information of the target user portrait; determine the similarity as the matching degree between the processing terminal and the breakpoint user terminal.

[0017] Furthermore, based on the matching degree and the behavior prediction value, the step of recommending user information to the processing terminal specifically includes:

[0018] Based on the behavior prediction value, the breakpoint user terminals are sorted to generate a breakpoint user priority sequence; based on the breakpoint user priority sequence, the priority information of the breakpoint user terminals is determined; based on the priority information and the matching degree, user information is recommended to the processing terminal.

[0019] 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:

[0020] A first data acquisition module, used to acquire user activity data of a user terminal in a target application;

[0021] A terminal determination module, used to determine a breakpoint user terminal from user terminals based on user activity data, and obtain breakpoint user information of the breakpoint user terminal;

[0022] A portrait acquisition module is used to acquire a target user portrait corresponding to a breakpoint user terminal based on a preset user portrait and breakpoint user information;

[0023] The prediction module is used to predict the behavior of the breakpoint user terminal based on the target user portrait and use a preset evaluation model to obtain a behavior prediction value;

[0024] A second data acquisition module is used to acquire historical sales data of the processing terminal, and calculate the matching degree between the processing terminal and the breakpoint user terminal based on the historical sales data and the target user portrait;

[0025] The recommendation module is used to recommend user information to the processing terminal based on the matching degree and behavior prediction value.

[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 embodiments of the present application mainly have the following beneficial effects: by acquiring and analyzing user active data, the user's behavior patterns and preferences in the target application can be more accurately understood, thereby providing more personalized services and product recommendations. By determining the breakpoint user terminal and obtaining its breakpoint user information, the user group that may be on the verge of churn or needs special attention can be accurately identified. By combining the preset user portrait and breakpoint user information to construct a target user portrait, the characteristics and needs of the breakpoint user can be more comprehensively understood. By using a preset evaluation model to predict the behavior of the breakpoint user terminal, a more accurate behavior prediction value can be obtained. Obtaining the historical sales data of the processing terminal and calculating the matching degree based on these data and the target user portrait can ensure the accuracy and effectiveness of the user information recommendation. This recommendation method based on data and algorithms can better meet the actual needs of users and improve the success rate of recommendations compared to traditional experience judgment or random recommendations. 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 a web page 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 1The 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 user activity data of a user terminal in a target application.

[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 user activity data of the user terminal in the target application 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] Among them, user terminals refer to the terminal devices used by users to access target applications (e.g., insurance apps, mini-programs and other online platforms), such as smartphones, tablets, desktop computers, notebooks, etc. User activity data generated by user activities in target applications can be collected through these devices.

[0047] The target application refers to the specific application or service that users access through user terminals (such as smartphones, tablets, desktop computers, notebooks, etc.), which can provide specific functions and content for users to use. For example, online platforms such as insurance apps and mini programs.

[0048] Among them, user activity data refers to various data related to the user's activity generated during the user's use of the target application (such as insurance APP, mini-programs and other online platforms). These data may include but are not limited to the user's login time, browsed pages, click behavior, stay time, search keywords, interactive operations and final transaction behavior. User activity data is an important basis for analyzing user behavior, understanding user needs and optimizing product services.

[0049] Step S202: determining a breakpoint user terminal from the user terminals based on the user activity data, and obtaining breakpoint user information of the breakpoint user terminal.

[0050] Among them, breakpoint user terminals refer to the terminal devices used by users who show specific behavior patterns in user activity data. Specifically, the "breakpoint" here refers to users who have left a certain number of activity footprints (such as browsing multiple pages, asking questions, etc.) on the target application (such as APP or mini program), but ultimately failed to complete the transaction or achieve a certain expected goal. These users may have interrupted the purchase or conversion process for various reasons (such as the need for further consideration, encountering operational obstacles, price factors, etc.).

[0051] Among them, breakpoint user information refers to a data set associated with the breakpoint user terminal that can further describe and identify these user characteristics. This information includes but is not limited to the user's basic information (such as gender, age, geographic location), historical behavior data (such as previous purchase records, consultation records), preference settings (such as browsing preferences, search keywords), and possible factors that hinder conversion (such as price sensitivity, trust issues, etc.). By collecting and analyzing this breakpoint user information, we can more accurately understand the needs and pain points of these users, so as to develop more effective service plans to improve conversion rates and user satisfaction.

[0052] Step S203: based on the preset user portrait and breakpoint user information, obtain the target user portrait corresponding to the breakpoint user terminal.

[0053] Among them, user portrait refers to a standardized user classification template built based on historical user data, that is, a series of representative user feature sets constructed based on historical user data obtained from the target application through data analysis and mining technology.

[0054] Among them, the target user portrait refers to a user portrait with personalized characteristics constructed for a specific breakpoint user terminal by matching and associating its breakpoint user information with a preset user portrait.

[0055] Step S204: Based on the target user portrait, a preset evaluation model is used to predict the behavior of the breakpoint user terminal to obtain a behavior prediction value.

[0056] The evaluation model refers to a series of pre-built and set algorithms or models used to analyze and predict user behavior. These models are usually based on machine learning, statistical analysis, or deep learning technologies, and can receive user profiles as input and output predictions about the future behavior of breakpoint users.

[0057] Behavior prediction refers to the process of using a preset evaluation model to predict the behavior of breakpoint users in the future. These behaviors may include but are not limited to: whether to complete the purchase, what product to buy, when to visit the app again, what types of marketing activities to respond positively, etc.

[0058] The behavior prediction value is the output of the behavior prediction process, which represents a quantitative estimate of the model's future behavior of the breakpoint user. The specific form of the prediction value depends on the design of the evaluation model and the prediction target. For example, if the prediction target is whether the breakpoint user will complete the purchase, then the prediction value can be a probability value between 0 and 1, indicating the probability that the user will complete the purchase. If the prediction target is the next time the breakpoint user visits the application, then the prediction value can be a specific time point or time period.

[0059] Step S205, obtaining historical sales data of the processing terminal, and calculating the matching degree between the processing terminal and the breakpoint user terminal based on the historical sales data and the target user portrait.

[0060] The processing terminal refers to the terminal devices used by insurance salesmen to perform daily work, such as smart phones, tablet computers or dedicated work computers. These devices are usually equipped with specific applications or software for managing user information, recording sales activities, receiving task assignments, etc.

[0061] Among them, historical sales data refers to the records of all sales activities completed by insurance salesmen through their processing terminals in the past period of time. These data may include information such as the types of insurance products sold, sales amount, sales time, customer feedback, conversion rate, etc.

[0062] Among them, the matching degree refers to the degree of adaptation between the processing terminal (i.e., insurance salesperson) and the breakpoint user terminal (i.e., potential customer) calculated based on a certain algorithm or model (e.g., cosine similarity). This matching degree usually takes into account multiple factors, such as the salesperson's sales expertise, historical sales performance, success stories similar to the breakpoint user, and the purchase intention and behavioral characteristics of the breakpoint user. By calculating the matching degree, it is possible to evaluate which salesperson is most suitable to follow up with which potential customer, thereby improving sales efficiency and conversion rate.

[0063] Step S206: Recommend user information to the processing terminal based on the matching degree and the behavior prediction value.

[0064] Among them, user information recommendation refers to the process of recommending appropriate breakpoint user information to the processing terminal (i.e., insurance salesperson) based on the matching degree and behavior prediction value. For example, in the intelligent allocation strategy, the system will automatically screen out the potential customers who are most likely to close a deal based on the characteristics of the salesperson and the needs of the potential customers (the breakpoint users corresponding to the breakpoint user terminal), and push the relevant information to the corresponding salesperson's processing terminal. This recommendation not only improves the work efficiency of the salesperson, but also makes sales activities more accurate and effective.

[0065] The embodiments of the present application can more accurately understand the user's behavior patterns and preferences in the target application by acquiring and analyzing user activity data, thereby providing more personalized services and product recommendations. By determining the breakpoint user terminal and obtaining its breakpoint user information, the user group that may be on the verge of churn or needs special attention can be accurately identified. By combining the preset user portrait and breakpoint user information to construct a target user portrait, the characteristics and needs of the breakpoint user can be more comprehensively understood. By using a preset evaluation model to predict the behavior of the breakpoint user terminal, a more accurate behavior prediction value can be obtained. Obtaining the historical sales data of the processing terminal and calculating the matching degree based on these data and the target user portrait can ensure the accuracy and effectiveness of the user information recommendation. This recommendation method based on data and algorithms can better meet the actual needs of users and improve the success rate of recommendations compared to traditional experience judgment or random recommendations.

[0066] In some optional implementations of this embodiment, step 202, determining a breakpoint user terminal from user terminals based on user activity data, specifically includes the following steps:

[0067] The user activity data is parsed to obtain the breakpoint data, user identification and behavior data of the user terminal; based on the breakpoint data, user identification and behavior data, a user status table of the user terminal is constructed; a preset behavior threshold is compared with the behavior data in the user status table to obtain a comparison result; based on the comparison result, it is determined whether the activity level of the user terminal reaches the behavior threshold; if the activity level does not reach the behavior threshold, the user terminal is marked as a breakpoint user terminal.

[0068] In one example, user active data can be collected through the user's login, browsing, clicking and other behaviors in the target application. The user active data is parsed using natural language processing technology to extract key information such as breakpoint data (such as the last active time), user identification (such as user ID, device ID) and behavior data (such as browsing time, browsing content, number of clicks) of the user terminal. Based on the extracted breakpoint data, user identification and behavior data, a user status table of the user terminal is constructed, which contains fields such as user ID, device ID, last active time, cumulative browsing time, browsing content, and cumulative number of clicks. Subsequently, the preset behavior threshold (such as the cumulative browsing time of less than 30 minutes for 7 consecutive days) is compared with the behavior data in the user status table, and the comparison result is obtained through a similarity calculation algorithm (such as Euclidean distance). The result ranges from 0 to 1, and the closer to 1, the less active it is. Finally, it is determined whether the activity level of the user terminal reaches the behavior threshold based on the comparison result. If the activity level does not reach the behavior threshold, the user terminal is marked as a breakpoint user terminal.

[0069] The embodiments of the present application can more accurately identify users who exhibit specific behavior patterns but fail to achieve their expected goals, i.e., breakpoint users, by analyzing user activity data in detail, including breakpoint data, user identification, and behavior data. By accurately identifying breakpoint users and implementing effective marketing strategies, the conversion rate of potential customers can be improved.

[0070] In some optional implementations of this embodiment, before obtaining the target user portrait corresponding to the breakpoint user terminal based on the preset user portrait and breakpoint user information in step S203, the following steps may also be included:

[0071] Obtain historical user data from the target application, preprocess the historical user data to obtain a processed user data set; extract features from the data in the user data set to obtain user feature information; and build a preset user portrait based on the user feature information.

[0072] In one example, historical user data for the past year can be obtained from the target application, including raw data such as user ID, device ID, login time, browsing time, and number of clicks. Then, the historical user data obtained is preprocessed, mainly including data cleaning, data integration, and data transformation. Data cleaning is mainly to remove noise data and abnormal data in the data, such as treating data with a login time of more than 24 hours as abnormal data and eliminating them; data integration is to integrate user data from different sources and in different formats to build a unified data storage; data transformation is to convert the format of time data and convert the timestamp into a standard date format. After preprocessing, a standardized and standard user data set is obtained. Next, feature extraction is performed on the data in the user data set to extract key features such as user activity, user preferences, and consumption capacity to form a structured user feature information table. Finally, a preset user portrait is constructed based on the user feature information table.

[0073] By constructing user portraits, the embodiments of the present application can more deeply understand key information such as user behavior patterns, preferences, and needs. Based on user portraits, personalized services and precision marketing can be achieved. For example, content and services that may be of interest to breakpoint users can be recommended. Such personalized services and marketing can significantly improve user experience and increase user satisfaction and loyalty.

[0074] In some optional implementations of this embodiment, the above step of “building a preset user portrait based on user feature information” specifically includes the following steps:

[0075] Match the user feature information with the preset user portrait template; if the user feature information matches the user portrait template, determine the user portrait template as the preset user portrait; if the user feature information does not match the user portrait template, use the portrait construction algorithm to construct the portrait based on the user feature information to generate the preset user portrait.

[0076] In one example, user feature information can be extracted from a user data set, mainly including user activity, user preference, consumption capacity, etc., and then the extracted user feature information is matched with a portrait template in a preset user portrait template library. Multiple typical user portraits can be preset in the user portrait template library. The cosine similarity algorithm is used to calculate the similarity between the user feature information and each portrait template. If the similarity is greater than a threshold, the portrait template is determined as a preset user portrait; if the similarity of all portrait templates is less than the threshold, it means that the breakpoint user corresponding to the breakpoint user terminal is a new user, and its features do not match the portrait template in the preset portrait library. It is necessary to use a portrait construction algorithm to construct a user portrait based on its feature information. Commonly used portrait construction algorithms include logistic regression, decision tree, neural network, etc. Taking the logistic regression algorithm as an example, the user feature information is used as input, the user portrait label is used as output, and a user portrait classification model is trained. When new user feature information is input into the model, the user portrait category to which it belongs can be predicted, and the portrait features under the category are used as its preset user portrait.

[0077] The embodiments of the present application can quickly perform preliminary screening and classification of a large amount of user feature information through a pre-set user portrait template. When the user features are highly matched with a certain template, the template is directly used as the user portrait, which can greatly improve the processing speed and accuracy and reduce the complex process of repeated calculations and portrait construction. When the user's feature information does not match the preset template, the portrait construction algorithm can be automatically triggered to dynamically generate a personalized user portrait based on the specific user feature information. This mechanism ensures the flexibility of the user portrait and its adaptability to complex user behaviors, making the user portrait closer to the real situation.

[0078] In some optional implementations of this embodiment, step S203, based on the preset user portrait and breakpoint user information, obtaining the target user portrait corresponding to the breakpoint user terminal, specifically includes the following steps:

[0079] Perform behavioral feature extraction on the breakpoint user information to obtain the behavioral feature information of the breakpoint user terminal; match the behavioral feature information with the feature label information of the user portrait to obtain a matching result; based on the matching result, determine the candidate user portrait corresponding to the breakpoint user terminal from the user portrait; associate the candidate user portrait with the behavioral feature information to generate a target user portrait of the breakpoint user terminal.

[0080] In one example, feature extraction can be performed on the breakpoint user information of the breakpoint user terminal. The behavioral feature information may include browsing history, search history, purchase history, access time, etc., and key features reflecting the interests and hobbies of the breakpoint user are extracted to obtain the behavioral feature information of the breakpoint user terminal. Then, the cosine similarity algorithm is used to calculate the similarity between the behavioral feature vector indicated by the behavioral feature information and the feature vector indicated by the feature label information of the user portrait. If the similarity is greater than the similarity threshold, it is considered that the user portrait matches the feature information of the breakpoint user, and the user portrait is determined as a candidate user portrait. The behavioral feature information of the breakpoint user terminal can be added as a new feature dimension to the feature label of the candidate user portrait to generate a personalized user portrait of the breakpoint user corresponding to the breakpoint user terminal, that is, to generate a target user portrait of the breakpoint user terminal. Through the above process, the real-time behavioral characteristics of the breakpoint user can be combined with the preset user portrait, and the user portrait can be dynamically updated to provide an accurate user preference basis for subsequent personalized recommendations.

[0081] The application embodiment can more accurately identify the current status and needs of breakpoint users (e.g., interrupted use, unfinished transactions, etc.) through behavioral feature extraction and matching. The target user portrait generated based on this information helps to provide more personalized services or recommendations, thereby improving user experience and increasing user satisfaction. The target user portrait provides a solid foundation for personalized recommendations and services.

[0082] In some optional implementations of this embodiment, step S205, based on the historical sales data and the target user portrait, calculating the matching degree between the processing terminal and the breakpoint user terminal, may specifically include the following steps:

[0083] Extract product attribute feature information of historical sales data; calculate the similarity between the product attribute feature information and the feature label information of the target user portrait; determine the similarity as the matching degree between the processing terminal and the breakpoint user terminal.

[0084] In one example, feature extraction can be performed on historical sales data to obtain product attribute feature information. Then, the Word2Vec algorithm can be used to map the extracted product attribute feature information into a vector representation, obtain the feature label vector of the feature label information in the target user portrait, and perform cosine similarity calculation on the feature vector obtained by mapping the product attribute feature information and the feature label vector to obtain the similarity between the product attribute feature information and the feature label information of the target user portrait. Among them, the feature label information of the target user portrait can be vectorized using the TF-IDF algorithm to obtain a feature label vector.

[0085] The embodiments of the present application can improve the accuracy of personalized recommendations, optimize user experience, and increase sales conversion rates by extracting product attribute feature information from historical sales data and calculating the similarity between the feature information and the target user portrait feature labels.

[0086] In some optional implementations of this embodiment, step S206, based on the matching degree and the behavior prediction value, recommending user information to the processing terminal, specifically includes the following steps:

[0087] Based on the behavior prediction value, the breakpoint user terminals are sorted to generate a breakpoint user priority sequence; based on the breakpoint user priority sequence, the priority information of the breakpoint user terminals is determined; based on the priority information and the matching degree, user information is recommended to the processing terminal.

[0088] In one example, the breakpoint user terminals can be sorted in descending order according to the behavior prediction value to generate a breakpoint user priority sequence. The higher the breakpoint user terminal ranks in the breakpoint user priority sequence, the higher its priority. For example, the priority sequence can be divided using the quantile method, and the top 20% of the breakpoint user terminals are defined as high priority, the breakpoint user terminals ranked between 20% and 50% are defined as medium priority, and the breakpoint user terminals ranked after 50% are defined as low priority, and the priority information corresponding to each breakpoint user is obtained, and the priority information includes the priority ranking and the priority score. Finally, the priority information and matching degree of the breakpoint user terminal are comprehensively considered, and the user information of the corresponding breakpoint user terminal is sent to the corresponding processing terminal to implement user information recommendation for the processing terminal.

[0089] The embodiment of the present application can sort breakpoint users by behavior prediction value, and can give priority to breakpoint user terminals that are more likely to generate positive behaviors. In this way, the processing terminal can be more accurate and efficient when allocating resources (such as recommending user information), thereby achieving optimal resource allocation. Prioritizing the recommendation of user information of high-priority users to processing terminals with high matching degrees can improve sales conversion rates.

[0090] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned user activity data, breakpoint user information, user portraits, target user portraits, behavior prediction values ​​and historical sales data, the above-mentioned user activity data, breakpoint user information, user portraits, target user portraits, behavior prediction values ​​and historical sales data can also be stored in a blockchain node.

[0091] 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.

[0092] The embodiments of the present application can acquire and process user activity data, breakpoint user information, user portraits, target user portraits, behavior prediction values, and historical sales data based on the relevant algorithms of natural language processing technology (NLP) under artificial intelligence. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0093] Natural language processing technology is a key branch in the field of artificial intelligence. It focuses on giving computers the ability to understand and process human language, so that they can parse, understand and generate natural language text like humans. By simulating some core mechanisms of human language processing, NLP technology can deeply process, analyze and interpret text data, extract key information, understand semantic content, and perform complex tasks or generate natural language responses based on it. Natural language processing technology can be used to analyze user activity data to determine breakpoint user terminals and obtain breakpoint user information of breakpoint user terminals. Based on user portraits and breakpoint user information, target user portraits of breakpoint users are obtained, and behavior predictions are made for breakpoint user terminals based on target user portraits. According to the behavior prediction values ​​obtained by behavior prediction and the historical sales data obtained, user information recommendations are made to processing terminals, etc.

[0094] 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).

[0095] 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.

[0096] 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.

[0097] like Figure 3 As shown, the information recommendation device 400 of this embodiment includes: a first data acquisition module 401, a terminal determination module 402, a portrait acquisition module 403, a prediction module 404, a second data acquisition module 405 and a recommendation module 406. Among them:

[0098] The first data acquisition module 401 is used to acquire user activity data of the user terminal in the target application;

[0099] The terminal determination module 402 is used to determine the breakpoint user terminal from the user terminals based on the user activity data, and obtain the breakpoint user information of the breakpoint user terminal;

[0100] The portrait acquisition module 403 is used to acquire the target user portrait corresponding to the breakpoint user terminal based on the preset user portrait and breakpoint user information;

[0101] The prediction module 404 is used to predict the behavior of the breakpoint user terminal based on the target user portrait using a preset evaluation model to obtain a behavior prediction value;

[0102] The second data acquisition module 405 is used to acquire the historical sales data of the processing terminal, and calculate the matching degree between the processing terminal and the breakpoint user terminal based on the historical sales data and the target user portrait;

[0103] The recommendation module 406 is used to recommend user information to the processing terminal based on the matching degree and the behavior prediction value.

[0104] The embodiments of the present application can more accurately understand the user's behavior patterns and preferences in the target application by acquiring and analyzing user activity data, thereby providing more personalized services and product recommendations. By determining the breakpoint user terminal and obtaining its breakpoint user information, the user group that may be on the verge of churn or needs special attention can be accurately identified. By combining the preset user portrait and breakpoint user information to construct a target user portrait, the characteristics and needs of the breakpoint user can be more comprehensively understood. By using a preset evaluation model to predict the behavior of the breakpoint user terminal, a more accurate behavior prediction value can be obtained. Obtaining the historical sales data of the processing terminal and calculating the matching degree based on these data and the target user portrait can ensure the accuracy and effectiveness of the user information recommendation. This recommendation method based on data and algorithms can better meet the actual needs of users and improve the success rate of recommendations compared to traditional experience judgment or random recommendations.

[0105] In one embodiment, the terminal determination module 402 includes:

[0106] The parsing submodule is used to parse the user activity data to obtain the breakpoint data, user identification and behavior data of the user terminal;

[0107] A construction submodule is used to construct a user status table of the user terminal based on the breakpoint data, the user identification and the behavior data;

[0108] A comparison submodule is used to compare the preset behavior threshold with the behavior data in the user status table to obtain a comparison result;

[0109] A judgment submodule is used to judge whether the activity level of the user terminal reaches the behavior threshold according to the comparison result;

[0110] The marking submodule is used to mark the user terminal as a breakpoint user terminal if the activity level does not reach the behavior threshold.

[0111] The embodiments of the present application can more accurately identify users who exhibit specific behavior patterns but fail to achieve their expected goals, i.e., breakpoint users, by analyzing user activity data in detail, including breakpoint data, user identification, and behavior data. By accurately identifying breakpoint users and implementing effective marketing strategies, the conversion rate of potential customers can be improved.

[0112] In one embodiment, the information recommendation device may further include:

[0113] A third data acquisition module is used to obtain historical user data from the target application, pre-process the historical user data, and obtain a processed user data set;

[0114] The feature extraction module is used to extract features from the data in the user data set to obtain user feature information;

[0115] The construction module is used to construct a preset user portrait based on user feature information.

[0116] By constructing user portraits, the embodiments of the present application can more deeply understand key information such as user behavior patterns, preferences, and needs. Based on user portraits, personalized services and precision marketing can be achieved. For example, content and services that may be of interest to breakpoint users can be recommended. Such personalized services and marketing can significantly improve user experience and increase user satisfaction and loyalty.

[0117] In one embodiment, a building block comprises:

[0118] The first matching submodule is used to match the user feature information with a preset user portrait template;

[0119] A first portrait determination submodule, configured to determine the user portrait template as a preset user portrait if the user feature information matches the user portrait template;

[0120] The portrait construction submodule is used to construct a portrait based on the user feature information using a portrait construction algorithm to generate a preset user portrait if the user feature information does not match the user portrait template.

[0121] The embodiment of the present application can quickly perform preliminary screening and classification of a large amount of user feature information through a pre-set user portrait template. When the user characteristics are highly matched with a certain template, the template is directly used as the user portrait, which can greatly improve the processing speed and accuracy and reduce the complex process of repeated calculations and portrait construction. When the user's feature information does not match the preset template, the portrait construction algorithm can be automatically triggered to dynamically generate a personalized user portrait based on the specific user feature information. This mechanism ensures the flexibility of the user portrait and its adaptability to complex user behaviors, making the user portrait closer to the real situation.

[0122] In one embodiment, the portrait acquisition module 403 includes:

[0123] The feature extraction submodule is used to extract the behavior feature of the breakpoint user information to obtain the behavior feature information of the breakpoint user terminal;

[0124] The second matching submodule is used to match the behavior feature information with the feature tag information of the user portrait to obtain a matching result;

[0125] A second portrait determination submodule is used to determine a candidate user portrait corresponding to the breakpoint user terminal from the user portraits based on the matching result;

[0126] The key submodule is used to associate the candidate user portrait with the behavioral feature information to generate the target user portrait of the breakpoint user terminal.

[0127] The embodiments of the present application can more accurately identify the current status and needs of breakpoint users (e.g., interrupted use, unfinished transactions, etc.) through behavioral feature extraction and matching. The target user portrait generated based on this information helps to provide more personalized services or recommendations, thereby improving user experience and increasing user satisfaction. The target user portrait provides a solid foundation for personalized recommendations and services.

[0128] In one embodiment, the second data acquisition module 405 includes:

[0129] An information extraction submodule is used to extract product attribute feature information from historical sales data;

[0130] A calculation submodule is used to calculate the similarity between the product attribute feature information and the feature label information of the target user portrait;

[0131] The matching degree determination submodule is used to determine the similarity as the matching degree between the processing terminal and the breakpoint user terminal.

[0132] The embodiment of the present application improves the accuracy of personalized recommendations, optimizes user experience, and increases sales conversion rate by extracting product attribute feature information from historical sales data and calculating the similarity between the feature information and the target user portrait feature label.

[0133] In one embodiment, the recommendation module 406 includes:

[0134] A sorting submodule is used to sort the breakpoint user terminals based on the behavior prediction value and generate a breakpoint user priority sequence;

[0135] An information determination submodule, used to determine the priority information of the breakpoint user terminal based on the breakpoint user priority sequence;

[0136] The recommendation submodule is used to recommend user information to the processing terminal based on priority information and matching degree.

[0137] The embodiment of the present application can sort breakpoint users by behavior prediction value, and can give priority to breakpoint user terminals that are more likely to generate positive behaviors. In this way, the processing terminal can be more accurate and efficient when allocating resources (such as recommending user information), thereby achieving optimal resource allocation. Prioritizing the recommendation of user information of high-priority users to processing terminals with high matching degrees can improve sales conversion rates.

[0138] 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.

[0139] 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.

[0140] 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.

[0141] 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.

[0142] 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.

[0143] 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.

[0144] The embodiments of the present application can more accurately understand the user's behavior patterns and preferences in the target application by acquiring and analyzing user activity data, thereby providing more personalized services and product recommendations. By determining the breakpoint user terminal and obtaining its breakpoint user information, the user group that may be on the verge of churn or needs special attention can be accurately identified. By combining the preset user portrait and breakpoint user information to construct a target user portrait, the characteristics and needs of the breakpoint user can be more comprehensively understood. By using a preset evaluation model to predict the behavior of the breakpoint user terminal, a more accurate behavior prediction value can be obtained. Obtaining the historical sales data of the processing terminal and calculating the matching degree based on these data and the target user portrait can ensure the accuracy and effectiveness of the user information recommendation. This recommendation method based on data and algorithms can better meet the actual needs of users and improve the success rate of recommendations compared to traditional experience judgment or random recommendations.

[0145] 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.

[0146] The embodiments of the present application can more accurately understand the user's behavior patterns and preferences in the target application by acquiring and analyzing user activity data, thereby providing more personalized services and product recommendations. By determining the breakpoint user terminal and obtaining its breakpoint user information, the user group that may be on the verge of churn or needs special attention can be accurately identified. By combining the preset user portrait and breakpoint user information to construct a target user portrait, the characteristics and needs of the breakpoint user can be more comprehensively understood. By using a preset evaluation model to predict the behavior of the breakpoint user terminal, a more accurate behavior prediction value can be obtained. Obtaining the historical sales data of the processing terminal and calculating the matching degree based on these data and the target user portrait can ensure the accuracy and effectiveness of the user information recommendation. This recommendation method based on data and algorithms can better meet the actual needs of users and improve the success rate of recommendations compared to traditional experience judgment or random recommendations.

[0147] 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.

[0148] 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 user activity data of the user terminal in the target application; Based on the user activity data, determining a breakpoint user terminal from the user terminals, and acquiring breakpoint user information of the breakpoint user terminal; Based on the preset user portrait and the breakpoint user information, obtaining a target user portrait corresponding to the breakpoint user terminal; Based on the target user portrait, a preset evaluation model is used to predict the behavior of the breakpoint user terminal to obtain a behavior prediction value; Acquire historical sales data of the processing terminal, and calculate the matching degree between the processing terminal and the breakpoint user terminal based on the historical sales data and the target user portrait; Based on the matching degree and the behavior prediction value, user information is recommended to the processing terminal.

2. The method according to claim 1, characterized in that The step of determining the breakpoint user terminal from the user terminals based on the user activity data specifically includes: Parsing the user activity data to obtain breakpoint data, user identification and behavior data of the user terminal; constructing a user status table of the user terminal based on the breakpoint data, the user identifier and the behavior data; Comparing the preset behavior threshold with the behavior data in the user status table to obtain a comparison result; Determining whether the activity level of the user terminal reaches the behavior threshold according to the comparison result; If the activity level does not reach the behavior threshold, the user terminal is marked as a breakpoint user terminal.

3. The method according to claim 1, characterized in that Before the step of acquiring the target user portrait corresponding to the breakpoint user terminal based on the preset user portrait and the breakpoint user information, the step further includes: Acquire historical user data from the target application, and preprocess the historical user data to obtain a processed user data set; Extracting features from the data in the user data set to obtain user feature information; A preset user portrait is constructed based on the user characteristic information.

4. The method according to claim 3, characterized in that The step of constructing a preset user portrait based on the user characteristic information specifically includes: Matching the user characteristic information with a preset user portrait template; If the user characteristic information matches the user portrait template, determining the user portrait template as a preset user portrait; If the user characteristic information does not match the user portrait template, a portrait construction algorithm is used to construct the portrait based on the user characteristic information to generate a preset user portrait.

5. The method according to claim 3, characterized in that: The step of acquiring the target user portrait corresponding to the breakpoint user terminal based on the preset user portrait and the breakpoint user information specifically includes: Extracting behavior characteristics of the breakpoint user information to obtain behavior characteristic information of the breakpoint user terminal; Matching the behavior feature information with the feature tag information of the user portrait to obtain a matching result; Based on the matching result, determining a candidate user profile corresponding to the breakpoint user terminal from the user profiles; The candidate user portrait is associated with the behavior feature information to generate a target user portrait of the breakpoint user terminal.

6. The method according to claim 1, characterized in that The step of calculating the matching degree between the processing terminal and the breakpoint user terminal based on the historical sales data and the target user portrait specifically includes: Extracting product attribute feature information of the historical sales data; Calculating the similarity between the product attribute feature information and the feature label information of the target user portrait; The similarity is determined as a matching degree between the processing terminal and the breakpoint user terminal.

7. The method according to claim 1, characterized in that The step of recommending user information to the processing terminal based on the matching degree and the behavior prediction value specifically includes: Based on the behavior prediction value, the breakpoint user terminals are sorted to generate a breakpoint user priority sequence; Based on the breakpoint user priority sequence, determining the priority information of the breakpoint user terminal; Based on the priority information and the matching degree, user information is recommended to the processing terminal.

8. An information recommendation device, characterized in that: include: A first data acquisition module, used to acquire user activity data of a user terminal in a target application; A terminal determination module, configured to determine a breakpoint user terminal from the user terminals based on the user activity data, and obtain breakpoint user information of the breakpoint user terminal; A portrait acquisition module, used to acquire a target user portrait corresponding to the breakpoint user terminal based on a preset user portrait and the breakpoint user information; A prediction module, configured to predict the behavior of the breakpoint user terminal based on the target user portrait and using a preset evaluation model to obtain a behavior prediction value; A second data acquisition module is used to acquire historical sales data of the processing terminal, and calculate the matching degree between the processing terminal and the breakpoint user terminal based on the historical sales data and the target user portrait; A recommendation module is used to recommend user information to the processing terminal based on the matching degree and the behavior prediction value.

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