User recommendation method and system
Through multi-dimensional data fusion technology, a personalized feature model is established and the matching degree with merchants is calculated, and users with high matching degree are recommended for merchants, which solves the problem of the lack of multi-dimensional data fusion analysis in the existing recommendation system, and achieves more accurate user matching and Promotional effect.
Patent Information
- Application Number
- CN202411602156.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-11
- Publication Date
- 2025-02-18
AI Technical Summary
The existing recommendation system lacks fusion analysis of multi-dimensional data, resulting in poor recommendation results.
A user recommendation method is proposed, by obtaining multi-dimensional user information data, pre-processing, standardizing processing and multi-dimensional data fusion, establishing a personalized feature model, and calculating the matching degree with the merchant, and recommending users with high matching degree to merchants.
It achieves more accurate user matching and recommendation, improves promotion effect, and better access to potential consumer users near merchants.
Smart Images
Figure CN119474543A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a user recommendation method and system. Background Art
[0002] With the development of smart phones and mobile Internet, businesses hope to reach nearby potential consumers more accurately.
[0003] However, most existing recommendation systems are based on a single data source, such as geographic location or user behavior, and lack the fusion analysis of multi-dimensional data, resulting in poor recommendation results. Summary of the invention
[0004] In view of this, this application proposes a user recommendation method and system, the specific scheme is as follows:
[0005] In the first part, the present application proposes a user recommendation method, which comprises:
[0006] Acquire multi-dimensional user information data and pre-process the user information data;
[0007] Standardizing the preprocessed user information data based on a preset standard, and extracting user feature data corresponding to different dimensions from the standardized user information data based on a preset rule;
[0008] After fusing the user feature data of different dimensions using a preset multi-dimensional data fusion algorithm, modeling and analyzing the fused user feature data based on a preset system model to establish a personalized feature model for the user;
[0009] The matching degree between the personalized feature model and the merchant is calculated, and based on a preset recommendation strategy, users corresponding to the matching degree reaching a preset value are recommended to the merchant.
[0010] In a specific embodiment, the user information data includes: interest data, and / or online behavior data, and / or geographic location data, and / or social media interaction data.
[0011] In a specific embodiment, the “preprocessing the user information data” includes:
[0012] Identify and clean the user information data to ensure the accuracy and completeness of the acquired user information data;
[0013] Performing time alignment on the user information data of different dimensions based on a preset time frame;
[0014] The user information data after time alignment is uniformly formatted based on a preset recognizable format.
[0015] In a specific embodiment, the standardization process includes: normalization process, and / or Z-score standardization process, and / or discretization process.
[0016] In a specific embodiment, the preset multi-dimensional data fusion algorithm includes an early fusion algorithm, and / or a late fusion algorithm, and / or a joint learning algorithm;
[0017] The “fusing the user feature data of different dimensions” includes:
[0018] The user feature data of different dimensions are spliced after feature engineering and standardization, and input into a preset unified model; or,
[0019] After training the user feature data of different dimensions through their corresponding models, the results of different models are integrated; or,
[0020] Based on a neural network, the user feature data of different dimensions are fused through a shared hidden layer.
[0021] In a specific embodiment, the “recommending users corresponding to the matching degree reaching a preset value to the merchant based on a preset recommendation strategy” further includes:
[0022] According to the personalized feature model corresponding to the user whose matching degree reaches a preset value, personalized product and service push is provided to the user.
[0023] In a specific embodiment, it also includes:
[0024] Filtering user dynamic behavior data from the user information data;
[0025] Monitoring the user's dynamic behavior data at a preset frequency, and if the user's dynamic behavior data changes, dynamically adjusting the personalized feature model for the user according to the change;
[0026] The matching degree between the adjusted personalized feature model and the merchant is recalculated, and based on a preset recommendation strategy, users corresponding to the matching degree reaching a preset value are recommended to the merchant.
[0027] In a specific embodiment, it also includes:
[0028] One or any multiple of the user information data, the user feature data, and the personalized feature model are associated with the user and then stored.
[0029] In the second part, this application proposes a user recommendation system, including:
[0030] An acquisition module, used to acquire multi-dimensional user information data and pre-process the user information data;
[0031] An extraction module, used to perform standardization processing on the pre-processed user information data based on a preset standard, and extract user feature data corresponding to different dimensions from the standardized user information data based on a preset rule;
[0032] A fusion module, used to fuse the user feature data of different dimensions using a preset multi-dimensional data fusion algorithm, and then model and analyze the fused user feature data based on a preset system model to establish a personalized feature model for the user;
[0033] The recommendation module is used to calculate the matching degree between the personalized feature model and the merchant, and recommend users corresponding to the matching degree reaching a preset value to the merchant based on a preset recommendation strategy.
[0034] In a specific embodiment, the “preprocessing the user information data” includes:
[0035] A cleaning module, used to identify and clean the user information data to ensure the accuracy and completeness of the acquired user information data;
[0036] An alignment module, used for performing time alignment on the user information data of different dimensions based on a preset time frame;
[0037] The processing module is used to uniformly format the user information data after time alignment based on a preset recognizable format.
[0038] Beneficial effects:
[0039] This application uses multi-dimensional data fusion technology to fuse multi-dimensional information data such as user interests, behavior data, and geographic location to generate a comprehensive user feature representation; and by calculating the matching degree between the personalized feature model of the user and the merchant, the merchant is recommended to the merchant for users with a matching degree that reaches a preset value, thus achieving more accurate user matching. Through accurate recommendations, accurate recommendations of potential users near the merchant are achieved, thus improving the promotion effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 is a flow chart of a user recommendation method according to an embodiment of the present application;
[0041] Figure 2 is a schematic diagram of a user recommendation system module in an embodiment of the present application;
[0042] Figure 3 This is a diagram of the user recommendation method architecture of an embodiment of the present application;
[0043] Figure 4 This is a user recommendation flow chart of an embodiment of the present application.
[0044] Figure numerals: 1 - acquisition module; 2 - extraction module; 3 - fusion module; 4 - recommendation module. DETAILED DESCRIPTION
[0045] Hereinafter, various embodiments disclosed in the present application will be described more fully. The present application may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments disclosed in the present application to the specific embodiments disclosed herein, but the present application should be understood to cover all adjustments, equivalents and / or alternatives falling within the spirit and scope of the various embodiments disclosed in the present application.
[0046] The terms used in the various embodiments disclosed in the present application are only used to describe the purpose of specific embodiments and are not intended to limit the various embodiments disclosed in the present application. As used herein, the singular form is intended to also include the plural form, unless the context clearly indicates otherwise. Unless otherwise specified, all terms used here (including technical terms and scientific terms) have the same meaning as the meaning generally understood by ordinary technicians in the field of various embodiments disclosed in the present application. The terms (such as the terms defined in the dictionary generally used) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized meaning or an overly formal meaning, unless clearly defined in the various embodiments disclosed in the present application.
[0047] Example 1
[0048] Example 1 of the present application discloses a user recommendation method. This embodiment uses multi-dimensional data fusion technology to fuse multi-dimensional information data such as user interests, behavior data, and geographic location to generate a comprehensive user feature representation; and by calculating the matching degree between the personalized feature model of the user and the merchant, the merchant is recommended to the user whose matching degree reaches a preset value, thereby achieving more accurate user matching. Through accurate recommendation, accurate recommendation of potential users near the merchant is achieved, which improves the promotion effect. The specific process is as follows: Figure 1 , Attachment Figure 3 and attached Figure 4 The specific plan is as follows:
[0049] A user recommendation method, as shown in the attached manual Figure 1 As shown, the method includes:
[0050] 101. Obtain multi-dimensional user information data and pre-process the user information data. In practical applications, user information data can be obtained through mobile devices or social platforms, including but not limited to basic information, location information, interest tags and consumption preferences. The collected data may include structured, semi-structured and unstructured data, such as: Structured data: user basic information, consumption history, location information, etc. Unstructured data: text data such as user comments and social media posts. Semi-structured data: data in formats such as XML and JSON.
[0051] Furthermore, the acquired user information data is preprocessed, including but not limited to cleaning data: removing missing values, noise data and redundant information to ensure the accuracy and completeness of the data. Time synchronization: aligning the time of data from different sources to ensure that data of different dimensions are analyzed within the same time frame. Format conversion: converting data from different sources into a unified format for subsequent processing.
[0052] 102. The pre-processed user information data is standardized based on preset standards, and user feature data corresponding to different dimensions are extracted from the standardized user information data based on preset rules. Due to the different sources of multidimensional data, the data may have different units, dimensions and distributions. Standardization is to convert all data to the same range or distribution for subsequent analysis. Specifically, standardization includes normalization: scaling the data to a certain range (such as 0 to 1), which is suitable for scenarios with large differences in feature dimensions. Z-score standardization: converting data into a standard normal distribution based on the mean and standard deviation, which is suitable for situations with different data distributions. Discretization: converting continuous data into discrete categories to facilitate processing of data of specific dimensions, such as dividing age data into multiple intervals.
[0053] Furthermore, by extracting, selecting, and combining data features of different dimensions, features that are helpful for analysis and modeling are generated.
[0054] First, extract useful features from the standardized user information data. And select the features that have the greatest impact on the analysis target to reduce the dimension. Common methods include information gain, chi-square test, principal component analysis (PCA), etc. In practical applications, data of different dimensions can also be combined to generate new features. For example, combining the user's interests and current geographical location to generate personalized consumption recommendation features.
[0055] 103. After fusing the user feature data of different dimensions using the preset multi-dimensional data fusion algorithm, the fused user feature data is modeled and analyzed based on the preset system model to establish a personalized feature model for the user. In practical applications, common modeling and analysis methods include:
[0056] Classification and regression models: Use classification or regression models to predict user behavior or consumption intentions through the fused multi-dimensional data; Cluster analysis: Cluster users according to interests, behaviors and other characteristics to help businesses discover potential patterns in user groups; Recommendation system model: Based on the multi-dimensional characteristics of users, use recommendation system algorithms such as collaborative filtering, matrix decomposition, and graph neural networks to provide businesses with accurate user recommendations.
[0057] 104. Calculate the matching degree between the personalized feature model and the merchant, and recommend users with a matching degree that reaches a preset value to the merchant based on a preset recommendation strategy. In practical applications, a deep neural network (DNN) model or a recommendation system built based on a graph neural network (GNN) can be used to perform multi-dimensional similarity matching between merchants and users. And based on the preset recommendation strategy, recommend users with a matching degree that reaches a preset value to the merchant.
[0058] In a specific embodiment, user information data includes: hobby data, and / or online behavior data, and / or geographic location data, and / or social media interaction data. As an important part of user information, hobby data records the user's preferences and interests in various fields in detail, including but not limited to reading, music, movies, sports, food, etc., and can intuitively reflect the user's personal taste and lifestyle. Online behavior data records the user's various behavior trajectories on the Internet, such as browsing web pages, searching keywords, shopping records, video viewing history, etc., which not only reveals the user's information needs and consumption habits, but also provides valuable basis for personalized recommendations and precision marketing; geographic location data records the user's real-time location, frequented places, movement trajectory and other information, which helps to analyze the user's travel habits, life circle and possible points of interest; social media interaction data covers the user's interactive behaviors such as posting, commenting, liking, sharing, etc. on social media platforms, as well as the social relationship network with other users. These data can reflect the user's social activity and influence, and provide strong support for business analysis and marketing.
[0059] In a specific embodiment, “preprocessing user information data” includes:
[0060] Identify and clean user information data to ensure the accuracy and completeness of the acquired user information data;
[0061] Time-align user information data of different dimensions based on a preset time frame;
[0062] The user information data after time alignment is uniformly formatted based on a preset recognizable format.
[0063] In this embodiment, the acquired user information data is preprocessed, including but not limited to cleaning the data: removing missing values, noise data and redundant information to ensure the accuracy and integrity of the data. Time synchronization: aligning the time of data from different sources to ensure that data of different dimensions are analyzed within the same time frame. Format conversion: converting data from different sources into a unified format for subsequent processing.
[0064] In a specific embodiment, the standardization process includes: normalization process, and / or Z-score standardization process, and / or discretization process.
[0065] In this embodiment, due to the different sources of multidimensional data, the data may have different units, dimensions and distributions. Standardization is to convert all data to the same range or distribution for subsequent analysis. Specifically, standardization processing includes normalization: scaling the data to a certain range (such as 0 to 1), which is suitable for scenarios with large differences in feature dimensions. Z-score standardization: converting data into a standard normal distribution based on the mean and standard deviation, which is suitable for situations with different data distributions. Discretization: converting continuous data into discrete categories to facilitate processing of data of specific dimensions, such as dividing age data into multiple intervals.
[0066] In a specific embodiment, the preset multi-dimensional data fusion algorithm includes an early fusion algorithm, and / or a late fusion algorithm, and / or a joint learning algorithm;
[0067] “Integrating user feature data of different dimensions” includes:
[0068] The user feature data of different dimensions are spliced after feature engineering and standardization, and input into the preset unified model; or,
[0069] After training the user feature data of different dimensions through their corresponding models, the results of different models are integrated; or,
[0070] Based on the neural network, user feature data of different dimensions are fused through a shared hidden layer.
[0071] In this embodiment, the preset multi-dimensional data fusion algorithm includes:
[0072] Early Fusion: Before model training, data features of different modalities are connected and integrated as input to the model. This is usually achieved through vector concatenation or feature concatenation. Late Fusion: First, data of different modalities are processed independently and learned using their respective models. Then, the output results of different models are fused to obtain the final prediction result. Joint Learning: By using a unified model structure (such as a deep neural network) to simultaneously learn data of different modalities, feature fusion is performed in the same model. Typical joint learning techniques include multimodal neural networks in deep learning.
[0073] Furthermore, depending on the application scenario, the data fusion model is also different. The following are some common data fusion models:
[0074] Feature-level fusion: Data of different dimensions are spliced after feature engineering and standardization, and then input into a unified model. This method is suitable for situations with a large number of multi-dimensional features and is often used in traditional machine learning models (such as support vector machines, decision trees, random forests, etc.).
[0075] Decision-level Fusion: Different data dimensions can train their own models and make decisions independently, and then the results of different models are fused through weighted averaging, voting, or Bayesian decision making. This method is suitable for scenarios where data sources are independent and have large structural differences.
[0076] Deep Learning Fusion: In deep learning, fusion can be achieved through neural networks, where input data of different dimensions are learned and represented through shared hidden layers. Common methods include convolutional neural networks (CNN), recurrent neural networks (RNN), graph neural networks (GNN), etc. They can process many different types of data, such as images, text, time series, and graph structured data.
[0077] In addition, in many application scenarios, data changes dynamically, especially when it comes to user behavior and geographic location, so time series data processing is very important: Time window analysis: divide the user's dynamic behavior data into time windows, and use the data in each time window as an independent analysis unit. Time series model fusion: Based on time series models (such as LSTM, GRU), historical data and real-time data are integrated to capture the dynamic change trend of data and generate more accurate recommendations.
[0078] In a specific embodiment, “recommending users corresponding to a preset matching degree to the merchant based on a preset recommendation strategy” further includes:
[0079] Based on the personalized feature model corresponding to the user whose matching degree reaches the preset value, personalized product and service push is provided to the user.
[0080] In this embodiment, it also includes: based on the personalized feature model corresponding to the user, the merchant can provide personalized product and service push to different users to further increase the user's participation and purchase willingness.
[0081] In a specific embodiment, it also includes:
[0082] Filter out user dynamic behavior data from user information data;
[0083] Monitor user dynamic behavior data at a preset frequency. If the user dynamic behavior data changes, dynamically adjust the personalized feature model for the user according to the change;
[0084] The matching degree between the adjusted personalized feature model and the merchant is recalculated, and users corresponding to the matching degree reaching the preset value are recommended to the merchant based on the preset recommendation strategy.
[0085] In this embodiment, the system can also filter out user dynamic behavior data from the user information data. These data can reflect the user's latest preferences and activity trends in real time, such as recent shopping tendencies, changes in browsing habits, etc. Then the system will continue to monitor these user dynamic behavior data at a preset frequency, such as daily, weekly, or based on specific event triggers. If it is found that the user's dynamic behavior data has changed significantly, such as the user suddenly showing a strong interest in a certain type of product, the system will dynamically adjust the personalized feature model for the user based on these changes.
[0086] The adjusted personalized feature model will be closer to the user's current status, and the system will then recalculate the matching degree between this updated model and each merchant. The matching degree is calculated based on a complex algorithm, taking into account multiple dimensions such as user characteristics, merchant attributes, historical interaction records, etc., aiming to find the merchant that best meets the user's current needs. Finally, the system will automatically recommend users whose matching degree reaches or exceeds the preset value to the merchant based on the preset recommendation strategy. This strategy may include giving priority to highly matching users, considering user activity, merchant preferences and other factors to ensure the effectiveness of the recommendation and the satisfaction of both parties.
[0087] In a specific embodiment, it also includes:
[0088] One or more of the user information data, user feature data, and personalized feature models are associated with the user and stored. This associated storage method ensures the consistency and traceability of the data, so that in the subsequent analysis, recommendation, or customer service process, the specific user and its related data can be quickly and accurately located.
[0089] Example 2
[0090] Embodiment 2 of the present application discloses a user recommendation system, which systematizes a user recommendation method in Embodiment 1 to make it more practical. The overall structure diagram of the user recommendation system is shown in the attached specification. Figure 2 The specific plan is as follows:
[0091] This embodiment proposes a user recommendation system, which includes:
[0092] Acquisition module 1, used to acquire multi-dimensional user information data and pre-process the user information data;
[0093] Extraction module 2, used to perform standardization processing on the pre-processed user information data based on a preset standard, and extract user feature data corresponding to different dimensions from the standardized user information data based on a preset rule;
[0094] Fusion module 3, used to fuse user feature data of different dimensions using a preset multi-dimensional data fusion algorithm, and then model and analyze the fused user feature data based on a preset system model to establish a personalized feature model for the user;
[0095] The recommendation module 4 is used to calculate the matching degree between the personalized feature model and the merchant, and recommend users corresponding to the matching degree reaching a preset value to the merchant based on a preset recommendation strategy.
[0096] In a specific embodiment, “preprocessing user information data” includes:
[0097] A cleaning module is used to identify and clean user information data to ensure the accuracy and completeness of the acquired user information data;
[0098] An alignment module, used to time-align user information data of different dimensions based on a preset time frame;
[0099] The processing module is used to uniformly format the user information data after time alignment based on a preset recognizable format.
[0100] In a specific embodiment, the user information data includes: interest data, and / or online behavior data, and / or geographic location data, and / or social media interaction data.
[0101] In a specific embodiment, the standardization process includes: normalization process, and / or Z-score standardization process, and / or discretization process.
[0102] In a specific embodiment, the preset multi-dimensional data fusion algorithm includes an early fusion algorithm, and / or a late fusion algorithm, and / or a joint learning algorithm;
[0103] “Integrating user feature data of different dimensions” includes:
[0104] The user feature data of different dimensions are spliced after feature engineering and standardization, and input into the preset unified model; or,
[0105] After training the user feature data of different dimensions through their corresponding models, the results of different models are integrated; or,
[0106] Based on the neural network, user feature data of different dimensions are fused through a shared hidden layer.
[0107] In a specific embodiment, “recommending users corresponding to a preset matching degree to the merchant based on a preset recommendation strategy” further includes:
[0108] Based on the personalized feature model corresponding to the user whose matching degree reaches the preset value, personalized product and service push is provided to the user.
[0109] In a specific embodiment, it also includes:
[0110] Filter out user dynamic behavior data from user information data;
[0111] Monitor user dynamic behavior data at a preset frequency. If the user dynamic behavior data changes, dynamically adjust the personalized feature model for the user according to the change;
[0112] The matching degree between the adjusted personalized feature model and the merchant is recalculated, and users corresponding to the matching degree reaching the preset value are recommended to the merchant based on the preset recommendation strategy.
[0113] In a specific embodiment, it also includes:
[0114] One or more of the user information data, user feature data, and personalized feature models are associated with the user and stored.
[0115] This embodiment proposes a user recommendation system, which systematizes the user recommendation method of Embodiment 1 to make it more practical.
[0116] This application uses multi-dimensional data fusion technology to fuse multi-dimensional information data such as user interests, behavior data, and geographic location to generate a comprehensive user feature representation; and by calculating the matching degree between the personalized feature model of the user and the merchant, the merchant is recommended to the merchant for users with a matching degree that reaches a preset value, thus achieving more accurate user matching. Through accurate recommendations, accurate recommendations of potential users near the merchant are achieved, thus improving the promotion effect.
[0117] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present invention. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario may be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or may be changed accordingly and located in one or more devices different from the present implementation scenario.
Claims
1. A user recommendation method, characterized in that: include: Acquire multi-dimensional user information data and pre-process the user information data; Standardizing the preprocessed user information data based on a preset standard, and extracting user feature data corresponding to different dimensions from the standardized user information data based on a preset rule; After fusing the user feature data of different dimensions using a preset multi-dimensional data fusion algorithm, modeling and analyzing the fused user feature data based on a preset system model to establish a personalized feature model for the user; The matching degree between the personalized feature model and the merchant is calculated, and based on a preset recommendation strategy, users corresponding to the matching degree reaching a preset value are recommended to the merchant.
2. The method according to claim 1, characterized in that The user information data includes: interest data, and / or online behavior data, and / or geographic location data, and / or social media interaction data.
3. The method according to claim 1, characterized in that The “preprocessing the user information data” includes: Identify and clean the user information data to ensure the accuracy and completeness of the acquired user information data; Performing time alignment on the user information data of different dimensions based on a preset time frame; The user information data after time alignment is uniformly formatted based on a preset recognizable format.
4. The method according to claim 1, characterized in that The standardization processing includes: normalization processing, and / or Z-score standardization processing, and / or discretization processing.
5. The method according to claim 1, characterized in that The preset multi-dimensional data fusion algorithm includes an early fusion algorithm, and / or a late fusion algorithm, and / or a joint learning algorithm; The “fusing the user feature data of different dimensions” includes: The user feature data of different dimensions are spliced after feature engineering and standardization, and input into a preset unified model; or, After training the user feature data of different dimensions through their corresponding models, the results of different models are integrated; or, Based on a neural network, the user feature data of different dimensions are fused through a shared hidden layer.
6. The method according to claim 1, characterized in that The "recommending users corresponding to the matching degree reaching the preset value to the merchant based on the preset recommendation strategy" also includes: According to the personalized feature model corresponding to the user whose matching degree reaches a preset value, personalized product and service push is provided to the user.
7. The method according to claim 1, characterized in that Also includes: Filtering user dynamic behavior data from the user information data; Monitoring the user's dynamic behavior data at a preset frequency, and if the user's dynamic behavior data changes, dynamically adjusting the personalized feature model for the user according to the change; The matching degree between the adjusted personalized feature model and the merchant is recalculated, and based on a preset recommendation strategy, users corresponding to the matching degree reaching a preset value are recommended to the merchant.
8. The method according to claim 1, characterized in that Also includes: One or any multiple of the user information data, the user feature data, and the personalized feature model are associated with the user and then stored.
9. A user recommendation system, characterized in that: For implementing the user recommendation method according to any one of claims 1 to 8, the system comprises: An acquisition module, used to acquire multi-dimensional user information data and pre-process the user information data; An extraction module, used to perform standardization processing on the pre-processed user information data based on a preset standard, and extract user feature data corresponding to different dimensions from the standardized user information data based on a preset rule; A fusion module, used to fuse the user feature data of different dimensions using a preset multi-dimensional data fusion algorithm, and then model and analyze the fused user feature data based on a preset system model to establish a personalized feature model for the user; The recommendation module is used to calculate the matching degree between the personalized feature model and the merchant, and recommend users corresponding to the matching degree reaching a preset value to the merchant based on a preset recommendation strategy.
10. The system according to claim 9, characterized in that The “preprocessing the user information data” includes: A cleaning module, used to identify and clean the user information data to ensure the accuracy and completeness of the acquired user information data; An alignment module, used for performing time alignment on the user information data of different dimensions based on a preset time frame; The processing module is used to uniformly format the user information data after time alignment based on a preset recognizable format.