Personalized Information Pushing via Correlation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing merchandise marketing websites face challenges in recommending sellers effectively, as they often prioritize popular products and familiar sellers, leading to low computational efficiency and accuracy, especially when users have not recently searched for keywords, requiring large volumes of search data and high computational loads.
Innovation Solution
A personalized information pushing method and device that retrieves network behavior data from user terminals, determines the degree of correlation between user and information providing terminals within a set time window, and pushes relevant information based on these correlations, using a system that includes a network server, user terminals, and information providing terminals connected via a network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If large volume of search data over long period is stored to provide recommendations to more users, then coverage of user recommendations is improved, but system memory requirements increase
Solution Approach 1:
The patent extracts only the essential elements needed for recommendation (user ID, terminal ID, information providing terminal ID, time stamp) from the complete search data, storing only these key fields in the behavior data table rather than the full search data, thereby reducing memory requirements while maintaining recommendation capability
Solution Approach 2:
The patent changes the parameter of data retention from storing complete search data to storing only extracted behavior data with specific parameters (user ID, terminal ID, information providing terminal ID, time stamp), transforming the data structure to achieve space efficiency
2Ease of operation
If extemporaneous keyword extraction and search operations are performed during user browsing, then real-time recommendations are provided, but computational load increases and efficiency decreases
Solution Approach 1:
The patent performs keyword extraction and search operations in advance, building the behavior data table with pre-extracted keywords and search results before user browsing occurs, so that during actual user interaction, only simple data retrieval and matching operations are needed, eliminating the need for extemporaneous computation
Solution Approach 2:
The patent implements a dynamic recommendation system where the behavior data table is continuously updated with new user behaviors, and the recommendation results are dynamically adjusted based on the pre-computed behavior data, allowing real-time adaptability without heavy computation during user interaction
3Adaptability or versatility
If analysis of all user behavior data is performed for each search operation, then comprehensive recommendations are generated, but computational accuracy decreases due to low efficiency
Solution Approach 1:
The patent extracts and stores only the most relevant behavior data elements (user ID, terminal ID, information providing terminal ID, time stamp, and pre-computed search results) in the behavior data table, filtering out redundant information, thereby improving both the comprehensiveness of recommendations and the accuracy of computational results by focusing on essential data
Data Source
AI summary
Embodiments of the present application relate to a personalized information pushing method, a personalized information pushing device, and a computer program product for pushing personalized information. A personalized information pushing method is provided. The method includes retrieving network behavior data related to access operations performed by user terminals, determining a numerical value of a degree of correlation between a user terminal and a plurality of information providing terminals having a correlation within a set time window based on the network behavior data, retrieving information providing terminals corresponding to a first user terminal to form a first data set based on the numerical values, retrieving information of the information providing terminals from the first data set to generate a first information, and pushing the first information to the first user terminal.