Search optimization method and system based on historical behavior data
By building a user tag database and content feature vector database, using machine learning models to predict search intentions and generate personalized recommendation lists, the problem of irrelevant results and interest sorting in existing search engines is solved, and the intelligence and personalization of search engines is improved.
Patent Information
- Application Number
- CN202510554474.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-29
- Publication Date
- 2025-07-29
AI Technical Summary
Existing search engines output irrelevant results when users enter too many keywords, and cannot be sorted according to user interest, which reduces the user experience.
User behavior data is collected through the data acquisition module, a user tag database and content feature vector database are constructed, and a machine learning model is used to extract keywords and predict search intentions, and a personalized recommendation list is generated based on user interest vectors and interaction matrix optimization.
It realizes accurate search results sorting, improves search quality and user experience, and provides convenient, efficient and personalized search services.
Smart Images
Figure CN120386934A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of marketing systems, and in particular to a search optimization method and system based on historical behavior data. Background Art
[0002] With the continuous development of information technology, the Internet has become an important part of people's lives. When people need to find various information, they only need to open a web page and enter relevant keywords in the search engine to collect relevant information, and search for the information they need in a relatively short time. With the continuous development of Internet technology and the continuous expansion of information, people's demand for online information is increasing.
[0003] However, when users use search engines to search for information, if they enter too many words, the search engine will easily output too many search results that are irrelevant to the content that the user is searching for. Moreover, the existing search results cannot be sorted according to the user's interest, requiring the user to frequently flip through the pages, which reduces the user experience.
[0004] Therefore, those skilled in the art provide a search optimization method and system based on historical behavior data to solve the problems raised in the above background technology. Summary of the Invention
[0005] In order to solve the above technical problems, the present invention provides a search optimization system based on historical behavior data, including a data acquisition module, a data analysis module and a search optimization module. The data acquisition module is used to collect user behavior data with the authorization of the platform user and obtain user behavior data. The data analysis module analyzes the user behavior information and constructs a corresponding database. The search optimization module is used to collect multi-channel feedback through a workbench to obtain user feedback data.
[0006] Preferably: the data acquisition module collects user input information through the intelligent terminal connected to it, extracts and organizes the user input information, and obtains a keyword sequence; performs word segmentation operations on all sentence information of the search requirement description information to obtain word units of the sentence information; obtains the word features of the word units, the sentence features of the word units in the corresponding sentence information, and the text features of the word units in the search requirement description information.
[0007] Preferably: the data acquisition module creates a machine learning model based on the acquired word features, sentence features and text features using a set number of analysis sentences based on the machine learning algorithm; based on the machine learning model, the word features, sentence features and text features of the word units in each sentence information are used to perform keyword extraction operations on each search requirement description information.
[0008] Preferably, the data analysis module summarizes the user's hobbies, fields, and search tendencies based on the user's historical input information and search information, detects and analyzes the user's behavior to obtain a user behavior sequence; constructs a user label database based on the user behavior sequence to obtain a user label database; extracts content information from the workbench and constructs a content feature vector database to obtain a content feature vector database; constructs a scenario data library based on the user label database and the content feature vector database to obtain a scenario data library.
[0009] Preferably, the data analysis module is used to optimize similarity propagation according to the user label database to obtain a user similarity matrix; extract user-content interactions according to the content feature matrix to obtain a user-content interaction matrix; assign weights to the user-content interaction matrix based on the scenario data library and the content feature vector database to obtain a weighted user-content interaction matrix; construct an association rule database based on the weighted user-content interaction matrix to obtain an association rule database.
[0010] Preferably, the data analysis module is used to capture the current scenario of platform users to obtain the current scenario label; sort the rule priorities according to the current scenario label and the association rule library to obtain a recommended rule sorting set; calculate the user interest vector according to the user label library to obtain a user interest vector; generate a personalized recommendation list according to the recommended rule sorting set and the user interest vector to obtain a personalized recommendation list.
[0011] Preferably, the data analysis module is used to extract the user behavior trajectory according to the user behavior sequence to obtain user behavior trajectory data; divide the user behavior trajectory data into time windows to obtain a user behavior time window data set; construct a frequent behavior pattern database based on the user behavior time window data set to obtain a frequent behavior pattern database; receive the current search term data of platform users in real time to obtain the current search term data; predict the search intention according to the current search term data and the frequent behavior pattern database to obtain the search intention prediction data.
[0012] Preferably, the search optimization module is used to collect multi-channel feedback through the workbench to obtain user feedback data; iteratively optimize the search data for the personalized recommendation list and the search intention prediction data according to the user feedback data to achieve precise search work.
[0013] A search optimization method based on historical behavior data includes the following steps:
[0014] Step 1: The data acquisition module is used to collect user behavior data with the authorization of the platform user and obtain user behavior data; collect user input information through the connected smart terminal, extract and organize the user input information, and obtain the keyword sequence;
[0015] Step 2: The data analysis module builds a scenario data library, an association rule database, a personalized recommendation list, and obtains search intent prediction data based on the user's historical input information and search information;
[0016] Step 3: The search optimization module is used to collect multi-channel feedback through the workbench to obtain user feedback data; based on the user feedback data, the personalized recommendation list and search intent prediction data are iteratively optimized to achieve accurate search work.
[0017] Preferably: in step 2, the data analysis module summarizes the user's interests, fields, and search tendencies based on the user's historical input information and search information, detects and analyzes the user's behavior, and obtains a user behavior sequence; constructs a user tag database based on the user behavior sequence to obtain a user tag database; extracts content information from the workbench and constructs a content feature vector database to obtain a content feature vector database; constructs a scenario data library based on the user tag database and the content feature vector database to obtain a scenario data library;
[0018] The data analysis module optimizes similarity propagation based on the user tag database to obtain a user similarity matrix; extracts user and content interactions based on the content feature matrix to obtain a user and content interaction matrix; assigns interaction matrix weights to the user and content interaction matrix based on the scenario data library and content feature vector database to obtain a user and content interaction weighted matrix; and constructs an association rule database based on the user and content interaction weighted matrix to obtain an association rule database.
[0019] The data analysis module captures the real-time scenes of platform users to obtain the current scene tags; it sorts the rules according to the current scene tags and the association rule database to obtain the recommendation rule sorting set; it calculates the user interest vector according to the user tag library to obtain the user interest vector; it generates a personalized recommendation list based on the recommendation rule sorting set and the user interest vector to obtain the personalized recommendation list;
[0020] The data analysis module extracts user behavior trajectories based on user behavior sequences to obtain user behavior trajectory data; divides the user behavior trajectory data into time windows to obtain a user behavior time window data set; constructs a frequent behavior pattern database based on the user behavior time window data set to obtain a frequent behavior pattern database; receives search terms from platform users in real time to obtain current search term data; and predicts search intent based on the current search term data and the frequent behavior pattern database to obtain search intent prediction data.
[0021] Technical effects and advantages of the present invention:
[0022] Compared with the prior art, the present invention has the following beneficial effects:
[0023] 1. The present invention analyzes the historical behavior data provided by the search engine platform, and the system can deeply understand the behavior patterns and needs of the target audience, so as to accurately select keywords to continuously improve the recommendation effect and search quality of the smart workbench search, making the system more intelligent and personalized, and providing users with a more convenient, efficient and satisfactory search experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 This is a flow chart of a search optimization method based on historical behavior data provided in an embodiment of the present application. DETAILED DESCRIPTION
[0025] The present invention will be described in further detail below with reference to specific embodiments. The embodiments of the present invention are provided for purposes of illustration and description and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments are chosen and described to better illustrate the principles and practical applications of the present invention and to enable those skilled in the art to understand the present invention and design various embodiments with various modifications suitable for specific applications.
[0026] Example 1
[0027] See also Figure 1 In this embodiment, a search optimization system based on historical behavior data is provided. A search optimization system based on historical behavior data includes a data acquisition module, a data analysis module and a search optimization module. The data acquisition module is used to collect user behavior data with the authorization of the platform user and obtain user behavior data. The data analysis module analyzes the user behavior information and builds a corresponding database. The search optimization module is used to collect multi-channel feedback through a workbench to obtain user feedback data.
[0028] The data acquisition module collects user input information through the intelligent terminal connected to it, extracts and organizes the user input information, and obtains a keyword sequence; performs word segmentation operations on all sentence information of the search demand description information to obtain word units of the sentence information; obtains the word features of the word units, the sentence features of the word units in the corresponding sentence information, and the text features of the word units in the search demand description information.
[0029] The data acquisition module creates a machine learning model based on the acquired word features, sentence features and text features using a set number of analysis sentences based on the machine learning algorithm; based on the machine learning model, the word features, sentence features and text features of the word units in each sentence information are used to perform keyword extraction operations on each search requirement description information.
[0030] The data analysis module summarizes the user's interests, fields, and search tendencies based on the user's historical input information and search information, detects and analyzes the user behavior, and obtains a user behavior sequence; constructs a user tag database based on the user behavior sequence to obtain a user tag database; extracts content information from the workbench and constructs a content feature vector database to obtain a content feature vector database; constructs a scenario data library based on the user tag database and the content feature vector database to obtain a scenario data library.
[0031] The data analysis module is used to optimize similarity propagation based on the user tag database to obtain a user similarity matrix; extract user and content interactions based on the content feature matrix to obtain a user and content interaction matrix; assign interaction matrix weights to the user and content interaction matrix based on the scenario data library and the content feature vector database to obtain a user and content interaction weighted matrix; and construct an association rule database based on the user and content interaction weighted matrix to obtain an association rule database.
[0032] The data analysis module is used to capture the real-time scenes of platform users and obtain the current scene labels; sort the rule priorities according to the current scene labels and the association rule library to obtain the recommendation rule sorting set; calculate the user interest vector according to the user tag library to obtain the user interest vector; generate a personalized recommendation list according to the recommendation rule sorting set and the user interest vector to obtain a personalized recommendation list.
[0033] The data analysis module is used to extract user behavior trajectories based on the user behavior sequence to obtain user behavior trajectory data; divide the user behavior trajectory data into time windows to obtain a user behavior time window data set; construct a frequent behavior pattern database for the user behavior time window data set to obtain a frequent behavior pattern database; receive the current search term data in real time for platform users; and predict the search intention based on the current search term data and the frequent behavior pattern database to obtain search intention prediction data.
[0034] The search optimization module is used to collect multi-channel feedback through the workbench to obtain user feedback data; and iteratively optimize the search data for the personalized recommendation list and the search intention prediction data according to the user feedback data to achieve accurate search work.
[0035] A search optimization method based on historical behavior data, characterized by including the following steps:
[0036] Step 1: The data acquisition module is used to collect user behavior data under the authorization of platform users to obtain user behavior data; collect user input information through the connected intelligent terminal, extract and organize the user input information, and obtain a keyword sequence.
[0037] In step 2, the data analysis module constructs a scenario data library, an association rule database, a personalized recommendation list, and obtains search intention prediction data based on the user's historical input information and search information.
[0038] Step 3: The search optimization module is used to collect multi-channel feedback through the workbench to obtain user feedback data; and iteratively optimize the search data for the personalized recommendation list and the search intention prediction data according to the user feedback data to achieve accurate search work.
[0039] In step 2, the data analysis module summarizes the user's hobbies, user fields, and user search tendencies based on the user's historical input information and search information, detects and analyzes the user behavior to obtain a user behavior sequence; constructs a user label database based on the user behavior sequence to obtain a user label database; extracts content information from the workbench and constructs a content feature vector database to obtain a content feature vector database; constructs a scenario data library based on the user label database and the content feature vector database to obtain a scenario data library.
[0040] The data analysis module optimizes similarity propagation according to the user tag database to obtain a user similarity matrix; extracts user-content interactions according to the content feature matrix to obtain a user-content interaction matrix; assigns weights to the user-content interaction matrix based on the scenario data library and the content feature vector database to obtain a weighted user-content interaction matrix; constructs an association rule database based on the weighted user-content interaction matrix to obtain an association rule database;
[0041] The data analysis module captures the real-time scenario of platform users to obtain the current scenario tags; sorts the rule priorities according to the current scenario tags and the association rule database to obtain a recommended rule sorting set; calculates the user interest vectors according to the user tag library to obtain user interest vectors; generates a personalized recommendation list according to the recommended rule sorting set and the user interest vectors to obtain a personalized recommendation list;
[0042] The data analysis module extracts the user behavior trajectory according to the user behavior sequence to obtain user behavior trajectory data; divides the user behavior trajectory data into time windows to obtain a user behavior time window data set; constructs a frequent behavior pattern database for the user behavior time window data set to obtain a frequent behavior pattern database; receives the current search term data for platform users in real time to obtain the current search term data; predicts the search intention according to the current search term data and the frequent behavior pattern database to obtain search intention prediction data.
[0043] Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art and related fields without creative efforts shall fall within the protection scope of the present invention. The structures, devices, and operation methods not specifically described and explained in the present invention shall be implemented by conventional means in the art without special instructions and limitations.
Claims
1. A search optimization system based on historical behavior data, characterized in that, It includes a data acquisition module, a data analysis module, and a search optimization module. The data acquisition module is used to collect user behavior data under the authorization of platform users to obtain user behavior data. The data analysis module analyzes user behavior information and constructs a corresponding database. The search optimization module is used to collect multi-channel feedback through a workbench to obtain user feedback data.
2. The search optimization method and system based on historical behavior data according to claim 1, characterized in that, The data acquisition module collects user input information through a connected intelligent terminal, extracts and organizes the user input information to obtain a keyword sequence; performs word segmentation on all statement information of the search requirement description information to obtain word units of the statement information; obtains the word features of the word units, the statement features of the word units in the corresponding statement information, and the text features of the word units in the search requirement description information.
3. A search optimization method and system based on historical behavior data according to claim 1, characterized in that The data acquisition module creates a machine learning model using a set number of analysis statements based on a machine learning algorithm according to the obtained word features, statement features, and text features; performs keyword extraction operations on each search requirement description information based on the machine learning model using the word features, statement features, and text features of the word units in each statement information.
4. A search optimization method and system based on historical behavior data according to claim 1, characterized in that The data analysis module summarizes the user's hobbies, user's field, and user's search tendency based on the user's historical input information and search information, detects and analyzes the user's behavior to obtain a user behavior sequence; constructs a user label database according to the user behavior sequence to obtain a user label database; extracts content information from the workbench and constructs a content feature vector database to obtain a content feature vector database; constructs a scenario data library according to the user label database and the content feature vector database to obtain a scenario data library.
5. A search optimization method and system based on historical behavior data according to claim 1, characterized in that The data analysis module is used to optimize similarity propagation according to the user label database to obtain a user similarity matrix; extract user and content interactions according to the content feature matrix to obtain a user and content interaction matrix; Assign weights to the user and content interaction matrix according to the scenario data library and the content feature vector database to obtain a weighted user and content interaction matrix; construct an association rule database according to the weighted user and content interaction matrix to obtain an association rule database.
6. A search optimization method and system based on historical behavior data according to claim 1, characterized in that, The data analysis module is used to capture the current scenario of platform users to obtain the current scenario label; sort the rule priorities according to the current scenario label and the association rule library to obtain a recommended rule sorting set; calculate the user interest vector according to the user label library; Generate a personalized recommendation list according to the recommended rule sorting set and the user interest vector to obtain a personalized recommendation list.
7. A search optimization method and system based on historical behavior data according to claim 1, characterized in that The data analysis module is used to extract user behavior trajectories based on the user behavior sequence to obtain user behavior trajectory data; divide the user behavior trajectory data into time windows to obtain a user behavior time window data set; construct a frequent behavior pattern database for the user behavior time window data set to obtain a frequent behavior pattern database; receive the current search term data in real time for platform users; and predict the search intent based on the current search term data and the frequent behavior pattern database to obtain search intent prediction data.
8. A search optimization method and system based on historical behavior data according to claim 1, characterized in that, The search optimization module is used to collect multi-channel feedback through the workbench to obtain user feedback data; Iteratively optimize the search data for the personalized recommendation list and the search intent prediction data according to the user feedback data to achieve accurate search work.
9. A search optimization method based on historical behavior data according to any one of claims 1-8, characterized in that, It includes the following steps: Step 1: The data acquisition module is used to collect user behavior data with the authorization of platform users to obtain user behavior data; collect user input information through the connected intelligent terminal, extract and organize the user input information, and obtain the keyword sequence; Step 2: The data analysis module constructs a scenario data library, an association rule database, a personalized recommendation list, and obtains search intent prediction data based on the user's historical input information and search information; Step 3: The search optimization module is used to collect multi-channel feedback through the workbench to obtain user feedback data; Iteratively optimize the search data for the personalized recommendation list and the search intent prediction data according to the user feedback data to achieve accurate search work.
10. A search optimization method based on historical behavior data according to claim 9, characterized in that, In Step 2, the data analysis module summarizes the user's interests, user fields, and search tendencies based on the user's historical input information and search information, detects and analyzes the user's behavior to obtain the user behavior sequence; constructs a user label database based on the user behavior sequence to obtain a user label database; extracts content information from the workbench and constructs a content feature vector database to obtain a content feature vector database; constructs a scenario data library based on the user label database and the content feature vector database to obtain a scenario data library; The data analysis module performs similarity propagation optimization based on the user label database to obtain a user similarity matrix; extracts user-content interactions based on the content feature matrix to obtain a user-content interaction matrix; Assign weights to the user-content interaction matrix based on the scenario data library and the content feature vector database to obtain a weighted user-content interaction matrix; construct an association rule database based on the weighted user-content interaction matrix to obtain an association rule database; The data analysis module captures the current scene for platform users to obtain the current scene label; sorts the rule priorities based on the current scene label and the association rule database to obtain a recommended rule sorting set; calculates the user interest vector based on the user label library to obtain a user interest vector; generates a personalized recommendation list based on the recommended rule sorting set and the user interest vector to obtain a personalized recommendation list; The data analysis module extracts the user behavior trajectory according to the user behavior sequence to obtain the user behavior trajectory data; Divide the user behavior trajectory data into time windows to obtain the user behavior time window data set; construct a frequent behavior pattern database for the user behavior time window data set to obtain the frequent behavior pattern database; receive the search terms of the platform users in real time to obtain the current search term data; predict the search intention according to the current search term data and the frequent behavior pattern database to obtain the search intention prediction data.