User Preference Analysis via Screen Content Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional user preference analysis methods on mobile devices are inefficient due to their inability to accurately distinguish between content of interest and noise, leading to unreliable preference analysis and wasteful resource usage, especially on small screens where users must scroll through unwanted content to find desired information.
Innovation Solution
A method and device that record and analyze content data displayed on the screen, including presentation time, to extract user preferences by focusing on actively viewed content, using natural language processing for text and image analysis to compute preferences based on display duration and time, thereby filtering out non-relevant 'noise' content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional user activity tracking methods are used to analyze all content on webpages, then content coverage is comprehensive, but analysis reliability deteriorates due to noise content
Solution Approach 1:
The patent extracts only the content that is actually displayed on the screen at the time of user interaction, rather than analyzing all content on the webpage. This extraction mechanism filters out noise content that is not visible to the user, thereby improving preference analysis reliability while maintaining relevant content coverage.
Solution Approach 2:
The patent performs preliminary action by recording which content is displayed on the screen before analyzing user preferences. This preliminary recording of display state allows the system to distinguish between relevant content (actually displayed) and noise content (not displayed), improving analysis reliability.
2Quantity of substance
If all webpage content is analyzed to determine user preferences, then content completeness is high, but time and resources are wasted analyzing noise content
Solution Approach 1:
The patent extracts only displayed content from the full webpage content for analysis. By taking out only the relevant portion (content actually displayed on screen), the system reduces analysis time and computational resources while maintaining completeness of user-interest content.
Solution Approach 2:
The patent applies partial action by analyzing only a subset of content (displayed content) rather than the entire content. This partial analysis approach reduces time and resource consumption while sufficient for accurate preference analysis, avoiding excessive action on irrelevant content.
3Adaptability or versatility
If conventional scraping methods are used to detect user interface type and installed applications, then device compatibility is achieved, but preference analysis accuracy is reduced
Solution Approach 1:
The patent enables the system to self-determine user preferences by analyzing actual displayed content and user interaction patterns, rather than relying on imprecise scraping methods. This self-service approach through content analysis and interaction tracking improves preference analysis accuracy while maintaining device compatibility.
Solution Approach 2:
The patent replaces the mechanical scraping method with a more sophisticated content analysis system that records displayed content and analyzes user interactions. This substitution improves measurement precision for preference analysis while maintaining adaptability across different devices.
4Weight of moving object
If small display screens are used in mobile devices, then device portability is improved, but content consumption efficiency deteriorates due to scrolling requirements
Solution Approach 1:
The patent introduces feedback by tracking user interactions with displayed content and using this information to improve preference analysis. This feedback mechanism allows the system to understand what content users actually engage with on small screens, improving content consumption efficiency despite limited display real estate.
Solution Approach 2:
The patent adds another dimension to content analysis by incorporating temporal information (when content is displayed) and interaction data into the preference analysis. This multi-dimensional approach compensates for the limitations of small displays by analyzing the sequence and context of content consumption.
Data Source
AI summary
A user preference analysis method and device are provided for providing a user with customized content through preference analysis based on the user's content consumption activity. The user preference analysis method includes displaying content data on a screen of the device in response to a user input; recording content information about the content data displayed on the screen; analyzing a user preference based on the recorded content information; and storing a user preference analysis result for the content data.