Information Filtering System Using Faceted Properties and Interest Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Internal networks within organizations fail to automatically identify and present relevant business information to employees, leading to employees seeking information manually and often missing important data.
Innovation Solution
A system that learns employee interests and narrows down information based on user-defined properties, inherent item characteristics, and computed properties to present the most relevant information, using techniques like faceting and interest graphs for information filtering and ranking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a comprehensive search system gathers information across the entire company network, then the quantity of available information increases, but the complexity of filtering and presenting relevant information increases
Solution Approach 1:
The patent segments information into discrete items with multiple properties (facets) such as author, department, date, and content tags. This segmentation allows the system to manage large quantities of information by breaking them into manageable units that can be independently filtered and organized, resolving the contradiction between information quantity and system complexity.
Solution Approach 2:
The patent introduces multiple dimensional properties (facets) for information items, transforming a flat information structure into a multi-dimensional space. Users can navigate and filter information along different dimensions (e.g., by department, by author, by date range), enabling efficient retrieval from large information sets without increasing system complexity.
2Loss of information
If the system presents all available information to users, then completeness of information is improved, but the ease of finding relevant information deteriorates
Solution Approach 1:
The system performs preliminary filtering and organization of information based on user profiles, historical behavior, and explicit preferences before presenting results. This preliminary action ensures that users receive complete yet relevant information without having to manually filter through all available data, improving both completeness and ease of finding information.
Solution Approach 2:
The system incorporates feedback mechanisms where user interactions (clicks, views, searches) are continuously analyzed to refine future information presentation. This feedback loop ensures that the system learns user preferences over time, maintaining information completeness while progressively improving ease of finding relevant information.
3Adaptability or versatility
If manual information seeking is required, then user control over search process is improved, but the loss of time increases
Solution Approach 1:
The system dynamically adjusts the balance between automated recommendations and user-controlled filtering. Users can start with automated personalized results and progressively apply their own filters and criteria, or vice versa. This dynamic approach maintains user control while significantly reducing the time required to find information compared to purely manual searching.
Data Source
AI summary
The disclosed technology provides systems and methods for filtering information based on a set of properties. The information consists of a set of items that the user is interacting with, such as documents, presentations, audio and video files, and the like. The properties can be specified by the user (by, for example, putting a set of items in lists and folders), based on actions taken by users in the system (such as commenting on, or liking, or viewing an item), or can represent a variety of other characteristics. Related properties can also be grouped together. Furthermore, the disclosed techniques provide mechanisms for automatically identifying useful properties and providing an indication of those useful properties to a user to use in narrowing results.