User behavior identification method and device for fast moving consumer goods and computer equipment

Abstract

The invention relates to a fast moving consumer goods user behavior identification method and device and computer equipment. The method comprises the steps of collecting user behavior data of each user of the fast moving consumer goods through a user behavior data platform with a multi-channel interface, and splitting the user behavior data into sub-behavior data of each user behavior type of each user; aiming at each user, respectively extracting behavior characteristics of each user behavior type, and aiming at each user behavior type, dividing each user into each user group; performing feature analysis processing on the behavior features of each user of each user group through a behavior feature analysis model to obtain user behavior information of each user group, and identifying user behavior distribution information of the user behavior type based on the user behavior information of each user group; and taking the user behavior distribution information of all the user behavior types as a target user behavior analysis result of the fast moving consumer goods. By adopting the method, the accuracy of user behavior analysis can be improved.

Description

Technical Field

[0001] The present application relates to the field of big data analysis technology, and in particular to a method, device and computer equipment for identifying user behavior of fast-moving consumer goods. Background Art

[0002] In the fast-moving consumer goods industry, understanding user behavior is of great significance to the formulation of business operation strategies, product function optimization, inventory management, etc. However, the amount of data corresponding to user behavior is huge, the data is scattered, the timeliness requirements are high, and the data acquisition cost is high. Therefore, how to improve the accuracy of user behavior analysis is the current research focus.

[0003] The traditional user behavior analysis method is to identify users' fast-moving consumer goods user behavior through consultation, manual collection, and manual analysis. However, the data obtained by this method is relatively one-sided, there are many subjective factors in manual analysis, and the timeliness is poor, resulting in poor accuracy in user behavior analysis. Summary of the invention

[0004] Based on this, it is necessary to provide a user behavior identification method, device, computer equipment, computer-readable storage medium and computer program product for fast-moving consumer goods in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for identifying user behavior of fast-moving consumer goods, comprising:

[0006] Collecting user behavior data of each user of the fast-moving consumer goods through a user behavior data platform with a multi-channel interface, and splitting each user behavior data into sub-behavior data of each user behavior type of each user;

[0007] For each user, based on the sub-behavior data of each user behavior type of the user, the behavior characteristics of each user behavior type are extracted respectively, and for each user behavior type, based on the behavior characteristics of each user of the user behavior type, the users are divided into user groups;

[0008] Performing feature analysis on the behavior features of each user in each user group through a behavior feature analysis model to obtain user behavior information of each user group, and identifying user behavior distribution information of the user behavior type based on the user behavior information of each user group;

[0009] The user behavior distribution information of all user behavior types is used as the target user behavior analysis result of the fast-moving consumer goods.

[0010] Optionally, the splitting of each of the user behavior data into sub-behavior data of each user behavior type of each user includes:

[0011] For each user behavior data, identify the behavior data content corresponding to each user behavior tag of the user, and query the user behavior type corresponding to each user behavior tag in the behavior database;

[0012] For each behavior data content, the behavior data content is split into actual user behavior content and online statistical data content, and based on the actual user behavior content and the online statistical data content, the target behavior content of the user is screened;

[0013] The target behavior content of each of the user behavior tags is used as sub-behavior data of the user behavior type corresponding to each of the user behavior tags.

[0014] Optionally, extracting the behavior features of each user behavior type based on the sub-behavior data of each user behavior type of the user includes:

[0015] Identify the data type of each sub-behavior data, and convert the sub-behavior data of each data type into each sub-text behavior data of a text type through a data conversion network of each data type;

[0016] Through the text feature extraction network, each sub-behavior feature of each sub-text behavior data is extracted respectively, and each sub-behavior feature of each sub-text behavior data is used as the behavior feature of the user behavior type corresponding to each sub-text behavior data.

[0017] Optionally, dividing the users into user groups based on the behavior characteristics of the users of the user behavior types includes:

[0018] By using a feature similarity recognition algorithm, the feature similarities between each sub-behavior feature of each user are calculated respectively, and the number of similar features of the sub-behavior features between the users that are greater than a feature similarity threshold is identified;

[0019] Based on the number of similar features between the users, clustering is performed on the users to obtain user groups of the user behavior type.

[0020] Optionally, the behavior characteristics of each user in each user group are analyzed and processed by the behavior characteristic analysis model to obtain user behavior information of each user group, including:

[0021] For each user group, the behavior characteristics of each user in the user group are subjected to feature vectorization processing to obtain a behavior feature vector of each user, and based on the behavior feature vector of each user, a behavior feature vector matrix of the user group is constructed;

[0022] Inputting the behavior feature vector matrix into a behavior feature analysis model, identifying each key feature vector point of the behavior feature vector, and querying the target behavior feature corresponding to each key feature vector point in the behavior feature vector of each user based on each key feature vector point;

[0023] The target behavior feature corresponding to each key feature vector point is used as the user behavior information of the user group.

[0024] Optionally, the identifying user behavior distribution information of the user behavior type based on the user behavior information of each user group includes:

[0025] Based on the target behavior characteristics of each user group, identify the user behavior preferences of each user group and the user behavior trends of each user group;

[0026] Based on the user behavior preferences of each of the user groups and the user behavior trends of each of the user groups, user behavior distribution information of the user behavior type is generated.

[0027] In a second aspect, the present application also provides a user behavior identification device for fast-moving consumer goods, including:

[0028] An acquisition module, for collecting user behavior data of each user of the fast-moving consumer goods through a user behavior data platform with a multi-channel interface, and splitting each user behavior data into sub-behavior data of each user behavior type of each user;

[0029] An extraction module, configured to extract, for each user, the behavior characteristics of each user behavior type based on the sub-behavior data of each user behavior type of the user, and, for each user behavior type, divide each user into user groups based on the behavior characteristics of each user of the user behavior type;

[0030] An identification module, used to perform feature analysis on the behavior features of each user in each user group through a behavior feature analysis model, obtain user behavior information of each user group, and identify user behavior distribution information of the user behavior type based on the user behavior information of each user group;

[0031] The determination module is used to use the user behavior distribution information of all user behavior types as the target user behavior analysis result of the fast-moving consumer goods.

[0032] Optionally, the acquisition module is specifically used to:

[0033] For each user behavior data, identify the behavior data content corresponding to each user behavior tag of the user, and query the user behavior type corresponding to each user behavior tag in the behavior database;

[0034] For each behavior data content, the behavior data content is split into actual user behavior content and online statistical data content, and based on the actual user behavior content and the online statistical data content, the target behavior content of the user is screened;

[0035] The target behavior content of each of the user behavior tags is used as sub-behavior data of the user behavior type corresponding to each of the user behavior tags.

[0036] Optionally, the extraction module is specifically used to:

[0037] Identify the data type of each sub-behavior data, and convert the sub-behavior data of each data type into each sub-text behavior data of a text type through a data conversion network of each data type;

[0038] Through the text feature extraction network, each sub-behavior feature of each sub-text behavior data is extracted respectively, and each sub-behavior feature of each sub-text behavior data is used as the behavior feature of the user behavior type corresponding to each sub-text behavior data.

[0039] Optionally, the extraction module is specifically used to:

[0040] By using a feature similarity recognition algorithm, the feature similarities between each sub-behavior feature of each user are calculated respectively, and the number of similar features of the sub-behavior features between the users that are greater than a feature similarity threshold is identified;

[0041] Based on the number of similar features between the users, clustering is performed on the users to obtain user groups of the user behavior type.

[0042] Optionally, the identification module is specifically used to:

[0043] For each user group, the behavior characteristics of each user in the user group are subjected to feature vectorization processing to obtain a behavior feature vector of each user, and based on the behavior feature vector of each user, a behavior feature vector matrix of the user group is constructed;

[0044] Inputting the behavior feature vector matrix into a behavior feature analysis model, identifying each key feature vector point of the behavior feature vector, and querying the target behavior feature corresponding to each key feature vector point in the behavior feature vector of each user based on each key feature vector point;

[0045] The target behavior feature corresponding to each key feature vector point is used as the user behavior information of the user group.

[0046] Optionally, the identification module is specifically used to:

[0047] Based on the target behavior characteristics of each user group, identify the user behavior preferences of each user group and the user behavior trends of each user group;

[0048] Based on the user behavior preferences of each of the user groups and the user behavior trends of each of the user groups, user behavior distribution information of the user behavior type is generated.

[0049] In a third aspect, the present application provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0050] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0051] In a fifth aspect, the present application provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0052] The above-mentioned user behavior identification method, device and computer equipment for fast-moving consumer goods collect user behavior data of each user of the fast-moving consumer goods through a user behavior data platform with a multi-channel interface, and split each of the user behavior data into sub-behavior data of each user behavior type of each user; for each user, based on the sub-behavior data of each user behavior type of the user, extract the behavior characteristics of each user behavior type respectively, and for each user behavior type, based on the behavior characteristics of each user of the user behavior type, divide each of the users into user groups; through a behavior characteristic analysis model, perform characteristic analysis processing on the behavior characteristics of each user in each user group to obtain user behavior information of each user group, and identify user behavior distribution information of the user behavior type based on the user behavior information of each user group; use the user behavior distribution information of all user behavior types as the target user behavior analysis result of the fast-moving consumer goods. This solution collects the behavior data of different users online through a user behavior data platform with a multi-channel interface, and then divides them into different user behavior types, and identifies the user behavior characteristics of each user behavior type respectively, so as to perform group characteristic analysis on each user. Then, based on the user behavior information of each user group, the user behavior distribution information of each user behavior type is identified, thereby identifying the target user behavior analysis results. After improving the diversity, comprehensiveness, real-time, and efficiency of user data collection, each user behavior type and each user group is refined to group and analyze the user's behavior characteristics by type, thereby improving the comprehensiveness, precision, and refinement of the analysis. Finally, after multi-angle and comprehensive group analysis, this solution comprehensively analyzes the user behavior characteristic distribution information of each user behavior type, thereby identifying the target user behavior analysis results of fast-moving consumer goods. Not only can it comprehensively and accurately analyze user behavior from different angles, but it can also identify the behavior analysis results of the entire user behavior through a comprehensive analysis, thereby effectively avoiding the analysis error problems of limitations and biases, and comprehensively improving the accuracy of user behavior analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 A schematic diagram of a process flow of a method for identifying user behavior of fast-moving consumer goods in one embodiment;

[0055] Figure 2 A flowchart of an example of user behavior recognition for fast-moving consumer goods in one embodiment;

[0056] Figure 3 It is a structural block diagram of a user behavior identification device for fast-moving consumer goods in one embodiment;

[0057] Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] The user behavior identification method for fast-moving consumer goods provided in the embodiment of the present application can be applied to the application environment of user analysis of fast-moving consumer goods. Among them, the method can be applied to a terminal, a server, or a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, etc. Among them, the terminal collects the behavior data of different users online through a user behavior data platform with a multi-channel interface, and then divides them into different user behavior types, and identifies the user behavior characteristics of each user behavior type respectively, so as to perform group feature analysis on each user. Then, by identifying the user behavior distribution information of each user behavior type based on the user behavior information of each user group, the target user behavior analysis result is identified. After improving the diversity, comprehensiveness, real-timeness, and efficiency of user data collection, the user behavior characteristics of each user behavior type and each user group are refined, so as to group and analyze the user behavior characteristics by type, thereby improving the comprehensiveness, precision, and refinement of the analysis. Finally, after the multi-angle and comprehensive group analysis, the scheme comprehensively analyzes the user behavior characteristic distribution information of each user behavior type, so as to identify the target user behavior analysis result of fast-moving consumer goods. Not only can it conduct a comprehensive and accurate analysis of user behavior from different angles, but it can also identify the behavioral analysis results of the entire user behavior through all-round analysis, thereby effectively avoiding the problems of limited and biased analysis errors and comprehensively improving the accuracy of user behavior analysis.

[0060] In an exemplary embodiment, Figure 1 As shown, a method for identifying user behavior of fast-moving consumer goods is provided, and the method is applied to a terminal as an example for explanation, including the following steps S101 to S104. Among them:

[0061] Step S101 , collecting user behavior data of each user of fast-moving consumer goods through a user behavior data platform with a multi-channel interface, and splitting each user behavior data into sub-behavior data of each user behavior type of each user.

[0062] In this embodiment, the terminal obtains the user behavior data of each user of each fast-moving consumer product by connecting to the user operation data collected in real time by the client of the user behavior data platform. The user behavior data platform can be a platform for statistics of each user's browsing, preference, selection, ordering, retrieval, collection, linking, sharing, commenting and other behaviors of fast-moving consumer products. The platform is supported by ID-Mapping technology and EUI model to achieve multi-terminal data unification and onlineization, unified data entry; hundreds of millions of data storage, support online query in seconds, solve the problem of difficult processing of massive data; real-time data reporting, more than 10 models (funnel, attribution, heat, etc.) platformization, and support visual analysis; including more than 200 collection points, a platform for full collection of user behavior data. Among them, each channel interface includes but is not limited to APP (Application, application software) interface, public account interface, mini program interface, PC (Personal Computer, personal computer) interface, H5 (HTML5, markup language for building interactive web content) interface, etc. Among them, fast-moving consumer goods can include products related to fast-moving consumer goods in many aspects and from all angles, or products related to derivative products involved in fast-moving consumer goods. The fast-moving consumer goods include but are not limited to food and beverages, daily necessities, cosmetics and beauty products, household products, medical and health products, pet food and supplies, baby products, etc. The user behavior data includes the actual user behavior data of each user behavior type, as well as online statistical data. Among them, user behavior types include but are not limited to browsing type, preference type, selection type, order type, search type, collection type, link type, sharing type, comment type and other user behavior types. Among them, the actual user behavior data is the actual operation content of the user, and the online statistical data includes the user operation data that is mistakenly collected due to reasons such as slow program response, user accidental touch, program abnormality, etc., as well as the collected data corresponding to the user's actual operation content.

[0063] Step S102 , for each user, based on the sub-behavior data of each user behavior type of the user, the behavior characteristics of each user behavior type are extracted respectively, and for each user behavior type, based on the behavior characteristics of each user of the user behavior type, each user is divided into user groups.

[0064] In this embodiment, the terminal extracts the behavior characteristics of each user behavior type based on the sub-behavior data of each user behavior type for each user. Among them, the behavior characteristics include the operation characteristics of each operation content. When extracting the operation characteristics of each operation content, it is necessary to first convert each operation content into text content, and then extract the content characteristics of the text content to obtain the operation characteristics of each operation content. The specific conversion process and text extraction process will be described in detail later. Among them, the operation content includes sub-behavior data of different data types, and the data type includes data types such as text type and code type. The sub-behavior data of the text type, for example, the request content when the user initiates a request to the background, such as the clicked commodity information, trigger time, and the error message returned after the click interruption. The code type is the sub-behavior data of the user clicking, browsing, playing, inputting, sliding, etc. at the flat point. The terminal can extract the behavior characteristics based on the sub-behavior data of the code type, such as the user click volume, user conversion rate, user jump volume, user behavior path, user attention information, etc. Behavior characteristics. Finally, the terminal divides each user into user groups based on the behavior characteristics of each user of the user behavior type for each user behavior type. The method of dividing each user group is a clustering method, and the specific division process will be described in detail later.

[0065] Step S103, using a behavior feature analysis model, performs feature analysis on the behavior features of each user in each user group to obtain user behavior information of each user group, and based on the user behavior information of each user group, identifies user behavior distribution information of user behavior types.

[0066] In this embodiment, the terminal performs feature analysis on the behavior features of each user in each user group through a behavior feature analysis model to obtain user behavior information of each user group, and identifies user behavior distribution information of user behavior types based on the user behavior information of each user group. The behavior feature analysis model is a feature analysis model. The user behavior information of each user behavior type includes the target behavior features of each user in each user group. The user behavior distribution information is the distribution information corresponding to the user preference information and user behavior trends of each user group. The specific identification process will be described in detail later.

[0067] Step S104: taking the user behavior distribution information of all user behavior types as the target user behavior analysis result of the fast-moving consumer goods.

[0068] In this embodiment, the terminal uses the user behavior distribution information of all user behavior types as the target user behavior analysis result of the fast-moving consumer goods.

[0069] Based on the above scheme, through the user behavior data platform with multi-channel interface, the behavior data of different users are collected online, and then divided into different user behavior types, and the user behavior characteristics of each user behavior type are identified respectively, so as to perform group characteristic analysis on each user. Then, by identifying the user behavior distribution information of each user behavior type based on the user behavior information of each user group, the target user behavior analysis results are identified. After improving the diversity, comprehensiveness, real-time and efficiency of user data collection, the user behavior characteristics are grouped and analyzed by type by refining each user behavior type and each user group, thereby improving the comprehensiveness, precision and refinement of the analysis. Finally, after multi-angle and comprehensive group analysis, this scheme comprehensively analyzes the user behavior characteristic distribution information of each user behavior type, so as to identify the target user behavior analysis results of fast-moving consumer goods. Not only can the user behavior be comprehensively and accurately analyzed from different angles, but also the behavior analysis results of the entire user behavior can be identified through all-round analysis, thereby effectively avoiding the analysis error problems of limitations and bias, and comprehensively improving the accuracy of the user behavior analysis.

[0070] Optionally, each user behavior data is split into sub-behavior data of each user behavior type of each user, including: for each user behavior data, identifying the behavior data content corresponding to each user behavior tag of the user, and querying the user behavior type corresponding to each user behavior tag in the behavior database; for each behavior data content, splitting the behavior data content into the user's actual behavior content and online statistical data content, and filtering the user's target behavior content based on the user's actual behavior content and the online statistical data content; using the target behavior content of each user behavior tag as the sub-behavior data of the user behavior type corresponding to each user behavior tag.

[0071] In this embodiment, the terminal identifies the behavior data content corresponding to each user behavior tag of the user for each user behavior data, and queries the user behavior type corresponding to each user behavior tag in the behavior database. Among them, each user behavior data includes each user behavior tag, and each user behavior tag is a user behavior tag obtained by data tagging the content of each behavior data. Among them, the user text tag can be, but is not limited to, a text tag, a color tag, a symbol tag, or a letter tag. The user behavior tag will be marked after each operation data in the user behavior data. For example, the user selected product A 4 times, browsed product B 5 times, and searched product C six times. The number of selections, browsing, and searches of each product is not in order. Then, each time the user browses product B, the terminal will mark the browsing type user behavior tag after recording the operation record of browsing product B. Each time the user selects product A, the terminal will mark the selection type user behavior tag after selecting the operation record of product A. Each time the user searches for product C, the terminal will mark the search type user behavior tag after the user searches for the operation record of product C. Among them, the operation record can be a record data of image data type, a record data of text data type, a record data of program code data type, etc.

[0072] For each behavior data content, the terminal splits the behavior data content into the user's actual behavior content and the online statistical data content. Among them, the user's actual behavior content is the user operation content corresponding to the user's actual interactive operation on the multi-channel terminal, and the online statistical data content is the user operation content recorded in the online background. Then, the terminal filters the user's target behavior content based on the user's actual behavior content and the online statistical data content. The screening method is to filter the target behavior content of each user behavior tag at the time point corresponding to the user operation content corresponding to the user interactive operation and the time point corresponding to the user operation content recorded in the online background, as the sub-behavior data of the user behavior type corresponding to each user behavior tag.

[0073] Based on the above solution, after labeling, data splitting is performed to improve the comprehensiveness and accuracy of behavioral data analysis for each user behavior type. Then, through data screening, interference data with abnormal operations is removed, improving the accuracy and comprehensiveness of behavioral data analysis.

[0074] Optionally, based on the sub-behavior data of each user behavior type of the user, behavioral features of each user behavior type are extracted respectively, including: identifying the data type of each sub-behavior data, and converting the sub-behavior data of each data type into sub-text behavior data of a text type through a data conversion network of each data type; extracting each sub-behavior feature of each sub-text behavior data through a text feature extraction network, and using each sub-behavior feature of each sub-text behavior data as the behavioral feature of the user behavior type corresponding to each sub-text behavior data.

[0075] In this embodiment, the terminal identifies the data type of each sub-behavior data, and converts the sub-behavior data of each data type into each sub-text behavior data of the text type through the data conversion network of each data type. Among them, the data conversion network of the image data type is a convolutional neural network based on the self-attention mechanism. The terminal identifies the image content of the image data type through the network, and uses the image content as the sub-text behavior data of the image data type. The data conversion network of the program code data type is to read the text content corresponding to the data content of the program code data type through the program code reading program to obtain the sub-text behavior data of the program code data type.

[0076] Then, the terminal extracts each sub-behavior feature of each sub-text behavior data through a text feature extraction network, and uses each sub-behavior feature of each sub-text behavior data as a behavior feature of the user behavior type corresponding to each sub-text behavior data. The text feature extraction network is a large language model based on natural language processing technology.

[0077] Based on the above solution, after text conversion, text feature extraction is performed, which not only improves the accuracy of extraction, but also improves the comprehensiveness of extraction.

[0078] Optionally, based on the behavioral characteristics of each user of the user behavior type, each user is divided into user groups, including: calculating the feature similarity between each sub-behavior feature of each user through a feature similarity recognition algorithm, and identifying the number of similar features of the sub-behavior features between each user that is greater than a feature similarity threshold; based on the number of similar features between each user, clustering each user to obtain user groups of the user behavior type.

[0079] In this embodiment, the terminal calculates the feature similarity between each sub-behavior feature of each user through a feature similarity recognition algorithm, and identifies the number of similar features of the sub-behavior features between each user that is greater than the feature similarity threshold. The feature similarity threshold is a domain value preset in the terminal. The terminal identifies the number of similar features between each sub-behavior feature of each two users that is greater than the feature similarity threshold, and then clusters each user based on the number of similar features between each user through a K-means algorithm to obtain user groups of user behavior types. The feature similarity recognition algorithm is a cosine similarity recognition algorithm.

[0080] Based on the above scheme, by means of feature similarity recognition, the number of similar features of sub-behavior features between users that are greater than the feature similarity threshold is identified respectively, so as to cluster each user, ensuring that each user in each user group after clustering is a user with similar features under the condition of the user behavior type, thereby ensuring the accuracy of analysis of users with different features.

[0081] Optionally, the behavioral characteristics of each user in each user group are subjected to characteristic analysis processing through a behavioral characteristic analysis model to obtain user behavior information of each user group, including: for each user group, the behavioral characteristics of each user in the user group are subjected to characteristic vectorization processing to obtain a behavioral characteristic vector of each user, and based on the behavioral characteristic vector of each user, a behavioral characteristic vector matrix of the user group is constructed; the behavioral characteristic vector matrix is ​​input into the behavioral characteristic analysis model to identify each key characteristic vector point of the behavioral characteristic vector, and based on each key characteristic vector point, in the behavioral characteristic vector of each user, a target behavioral characteristic corresponding to each key characteristic vector point is queried; and the target behavioral characteristic corresponding to each key characteristic vector point is used as the user behavior information of the user group.

[0082] In this embodiment, the terminal performs feature vectorization processing on the behavior characteristics of each user in the user group for each user group, obtains the behavior feature vector of each user, and constructs the behavior feature vector matrix of the user group based on the behavior feature vector of each user. Among them, the feature vectorization processing method is to convert the behavior characteristics of each text data type into a feature vector by the feature vectorization method corresponding to the word embedding method. Then, the terminal inputs the behavior feature vector matrix into the behavior feature analysis model, identifies each key feature vector point of the behavior feature vector, and the matrix characteristics of the behavior feature vector matrix, and based on each key feature vector point, queries the target behavior feature corresponding to each key feature vector point in the behavior feature vector of each user. Among them, the behavior feature analysis model is an analysis model for performing matrix feature analysis on the feature matrix, and the model can identify the key feature vector points between each vector of the behavior feature vector matrix. Among them, the key feature vector point is a vector point in a feature vector, and the number of vector points with similar features as the vector points in other feature vectors is greater than the number threshold preset in the terminal, then the behavior feature corresponding to the vector point is the target behavior feature.

[0083] The terminal uses the target behavior feature corresponding to each key feature vector point as the user behavior information of the user group, wherein the target behavior feature is the behavior feature of each user that has a high similarity with other users.

[0084] Based on the above scheme, after feature vectorization processing, the target behavior features corresponding to the key feature vector points in the behavior features are identified through the feature analysis model, thereby improving the recognition efficiency and recognition accuracy.

[0085] Optionally, based on the user behavior information of each user group, user behavior distribution information of user behavior types is identified, including: based on the target behavior characteristics of each user group, user behavior preferences of each user group and user behavior trends of each user group are identified; based on the user behavior preferences of each user group and user behavior trends of each user group, user behavior distribution information of user behavior types is generated.

[0086] In this embodiment, the terminal identifies the user behavior preference of each user group and the user behavior trend of each user group based on each target behavior feature of each user group. The behavior content corresponding to each target behavior feature is the user behavior preference of each user group, and then the behavior content after each target behavior feature is the user behavior trend of each user group. For example, the target behavior feature is to select fast-moving consumer goods, and the user behavior preference is to purchase fast-moving consumer goods, compare fast-moving consumer goods, browse fast-moving consumer goods, collect fast-moving consumer goods, etc. Finally, the terminal generates user behavior distribution information of user behavior types based on the user behavior preferences of each user group and the user behavior trends of each user group.

[0087] Based on the above solution, after multi-angle and comprehensive group analysis, the user behavior characteristic distribution information of each user behavior type is comprehensively analyzed to identify the target user behavior analysis results of fast-moving consumer goods. Not only can the user behavior be comprehensively and accurately analyzed from different angles, but also the behavior analysis results of the entire user behavior can be identified through all-round analysis, thereby effectively avoiding the analysis errors caused by limitations and deviations, and comprehensively improving the accuracy of the user behavior analysis.

[0088] This application also provides an example of user behavior identification for fast-moving consumer goods, such as Figure 2 As shown, the specific processing process includes the following steps:

[0089] Step S201, collecting user behavior data of each user of fast-moving consumer goods through a user behavior data platform with a multi-channel interface.

[0090] Step S202: for each user behavior data, identifying the behavior data content corresponding to each user behavior tag of the user, and querying the user behavior type corresponding to each user behavior tag in the behavior database.

[0091] Step S203, for each behavior data content, split the behavior data content into user actual behavior content and online statistical data content, and filter the user's target behavior content based on the user's actual behavior content and the online statistical data content.

[0092] Step S204: taking the target behavior content of each user behavior tag as sub-behavior data of the user behavior type corresponding to each user behavior tag.

[0093] Step S205, identifying the data type of each sub-behavior data, and converting the sub-behavior data of each data type into each sub-text behavior data of a text type through a data conversion network of each data type.

[0094] Step S206, extracting the sub-behavior features of each sub-text behavior data through the text feature extraction network, and using the sub-behavior features of each sub-text behavior data as the behavior features of the user behavior type corresponding to each sub-text behavior data.

[0095] Step S207 , using a feature similarity recognition algorithm, respectively calculate the feature similarities between the sub-behavior features of each user, and identify the number of similar features of the sub-behavior features between the users that are greater than a feature similarity threshold.

[0096] Step S208: clustering the users based on the number of similar features between the users to obtain user groups of user behavior types.

[0097] Step S209 , for each user group, the behavior characteristics of each user in the user group are subjected to feature vectorization processing to obtain a behavior feature vector of each user, and based on the behavior feature vector of each user, a behavior feature vector matrix of the user group is constructed.

[0098] Step S210, inputting the behavior feature vector matrix into the behavior feature analysis model, identifying each key feature vector point of the behavior feature vector, and based on each key feature vector point, querying the target behavior feature corresponding to each key feature vector point in the behavior feature vector of each user.

[0099] Step S211: taking the target behavior feature corresponding to each key feature vector point as the user behavior information of the user group.

[0100] Step S212: identifying the user behavior preference of each user group and the user behavior trend of each user group based on the target behavior characteristics of each user group.

[0101] Step S213: generating user behavior distribution information of user behavior types based on the user behavior preferences of each user group and the user behavior trends of each user group.

[0102] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0103] Based on the same inventive concept, the embodiment of the present application also provides a user behavior identification device for fast-moving consumer goods for implementing the user behavior identification method for fast-moving consumer goods involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more fast-moving consumer goods user behavior identification devices provided below can refer to the limitations of the user behavior identification method for fast-moving consumer goods above, and will not be repeated here.

[0104] In an exemplary embodiment, Figure 3 As shown, a user behavior identification device for fast-moving consumer goods is provided, including: an acquisition module 310, an extraction module 320, an identification module 330 and a determination module 340, wherein:

[0105] An acquisition module 310 is used to collect user behavior data of each user of the fast-moving consumer goods through a user behavior data platform with a multi-channel interface, and split each user behavior data into sub-behavior data of each user behavior type of each user;

[0106] An extraction module 320 is used to extract, for each user, the behavior characteristics of each user behavior type based on the sub-behavior data of each user behavior type of the user, and, for each user behavior type, divide each user into user groups based on the behavior characteristics of each user of the user behavior type;

[0107] The identification module 330 is used to perform feature analysis on the behavior features of each user in each user group through a behavior feature analysis model to obtain user behavior information of each user group, and identify user behavior distribution information of the user behavior type based on the user behavior information of each user group;

[0108] The determination module 340 is used to use the user behavior distribution information of all user behavior types as the target user behavior analysis result of the fast-moving consumer goods.

[0109] Optionally, the acquisition module 310 is specifically configured to:

[0110] For each user behavior data, identify the behavior data content corresponding to each user behavior tag of the user, and query the user behavior type corresponding to each user behavior tag in the behavior database;

[0111] For each behavior data content, the behavior data content is split into actual user behavior content and online statistical data content, and based on the actual user behavior content and the online statistical data content, the target behavior content of the user is screened;

[0112] The target behavior content of each of the user behavior tags is used as sub-behavior data of the user behavior type corresponding to each of the user behavior tags.

[0113] Optionally, the extraction module 320 is specifically used to:

[0114] Identify the data type of each sub-behavior data, and convert the sub-behavior data of each data type into each sub-text behavior data of a text type through a data conversion network of each data type;

[0115] Through the text feature extraction network, each sub-behavior feature of each sub-text behavior data is extracted respectively, and each sub-behavior feature of each sub-text behavior data is used as the behavior feature of the user behavior type corresponding to each sub-text behavior data.

[0116] Optionally, the extraction module 320 is specifically used to:

[0117] By using a feature similarity recognition algorithm, the feature similarities between each sub-behavior feature of each user are calculated respectively, and the number of similar features of the sub-behavior features between the users that are greater than a feature similarity threshold is identified;

[0118] Based on the number of similar features between the users, clustering is performed on the users to obtain user groups of the user behavior type.

[0119] Optionally, the identification module 330 is specifically used to:

[0120] For each user group, the behavior characteristics of each user in the user group are subjected to feature vectorization processing to obtain a behavior feature vector of each user, and based on the behavior feature vector of each user, a behavior feature vector matrix of the user group is constructed;

[0121] Inputting the behavior feature vector matrix into a behavior feature analysis model, identifying each key feature vector point of the behavior feature vector, and querying the target behavior feature corresponding to each key feature vector point in the behavior feature vector of each user based on each key feature vector point;

[0122] The target behavior feature corresponding to each key feature vector point is used as the user behavior information of the user group.

[0123] Optionally, the identification module 330 is specifically used to:

[0124] Based on the target behavior characteristics of each user group, identify the user behavior preferences of each user group and the user behavior trends of each user group;

[0125] Based on the user behavior preferences of each of the user groups and the user behavior trends of each of the user groups, user behavior distribution information of the user behavior type is generated.

[0126] Each module in the above-mentioned user behavior identification device for fast-moving consumer goods can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0127] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 4As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a user behavior identification method for fast-moving consumer goods is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0128] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0129] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0130] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0131] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of any one of the methods in the first aspect.

[0132] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0133] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0134] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for identifying user behavior of fast-moving consumer goods, characterized in that: The method comprises: Collecting user behavior data of each user of the fast-moving consumer goods through a user behavior data platform with a multi-channel interface, and splitting each user behavior data into sub-behavior data of each user behavior type of each user; For each user, based on the sub-behavior data of each user behavior type of the user, the behavior characteristics of each user behavior type are extracted respectively, and for each user behavior type, based on the behavior characteristics of each user of the user behavior type, the users are divided into user groups; Performing feature analysis on the behavior features of each user in each user group through a behavior feature analysis model to obtain user behavior information of each user group, and identifying user behavior distribution information of the user behavior type based on the user behavior information of each user group; The user behavior distribution information of all user behavior types is used as the target user behavior analysis result of the fast-moving consumer goods.

2. The method according to claim 1, characterized in that The user behavior data is split into sub-behavior data of each user behavior type of each user, including: For each user behavior data, identify the behavior data content corresponding to each user behavior tag of the user, and query the user behavior type corresponding to each user behavior tag in the behavior database; For each behavior data content, the behavior data content is split into actual user behavior content and online statistical data content, and based on the actual user behavior content and the online statistical data content, the target behavior content of the user is screened; The target behavior content of each of the user behavior tags is used as sub-behavior data of the user behavior type corresponding to each of the user behavior tags.

3. The method according to claim 2, characterized in that The extracting behavior features of each user behavior type based on the sub-behavior data of each user behavior type of the user includes: Identify the data type of each sub-behavior data, and convert the sub-behavior data of each data type into each sub-text behavior data of a text type through a data conversion network of each data type; Through the text feature extraction network, each sub-behavior feature of each sub-text behavior data is extracted respectively, and each sub-behavior feature of each sub-text behavior data is used as the behavior feature of the user behavior type corresponding to each sub-text behavior data.

4. The method according to claim 3, characterized in that: The dividing the users into user groups based on the behavior characteristics of the users of the user behavior type includes: By using a feature similarity recognition algorithm, the feature similarities between each sub-behavior feature of each user are calculated respectively, and the number of similar features of the sub-behavior features between the users that are greater than a feature similarity threshold is identified; Based on the number of similar features between the users, clustering is performed on the users to obtain user groups of the user behavior type.

5. The method according to claim 4, characterized in that The behavior feature analysis model is used to analyze the behavior features of each user in each user group to obtain user behavior information of each user group, including: For each user group, the behavior characteristics of each user in the user group are subjected to feature vectorization processing to obtain a behavior feature vector of each user, and based on the behavior feature vector of each user, a behavior feature vector matrix of the user group is constructed; Inputting the behavior feature vector matrix into a behavior feature analysis model, identifying each key feature vector point of the behavior feature vector, and querying the target behavior feature corresponding to each key feature vector point in the behavior feature vector of each user based on each key feature vector point; The target behavior feature corresponding to each key feature vector point is used as the user behavior information of the user group.

6. The method according to claim 5, characterized in that The identifying the user behavior distribution information of the user behavior type based on the user behavior information of each user group includes: Based on the target behavior characteristics of each user group, identify the user behavior preferences of each user group and the user behavior trends of each user group; Based on the user behavior preferences of each of the user groups and the user behavior trends of each of the user groups, user behavior distribution information of the user behavior type is generated.

7. A user behavior recognition device for fast-moving consumer goods, characterized in that: The device comprises: An acquisition module, for collecting user behavior data of each user of the fast-moving consumer goods through a user behavior data platform with a multi-channel interface, and splitting each user behavior data into sub-behavior data of each user behavior type of each user; An extraction module, configured to extract, for each user, the behavior characteristics of each user behavior type based on the sub-behavior data of each user behavior type of the user, and, for each user behavior type, divide each user into user groups based on the behavior characteristics of each user of the user behavior type; An identification module, used to perform feature analysis on the behavior features of each user in each user group through a behavior feature analysis model, obtain user behavior information of each user group, and identify user behavior distribution information of the user behavior type based on the user behavior information of each user group; The determination module is used to use the user behavior distribution information of all user behavior types as the target user behavior analysis result of the fast-moving consumer goods.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images