Digital Content Credibility Scoring via Multi-Level Usage Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In closed and secure environments, it is challenging for internal search engines to surface the relevance or quality of documents due to a smaller user base and continuously streaming content, making it difficult to prioritize documents that are more important to search queries.
Innovation Solution
A method that assigns a usage score to each digital content item based on different levels of user interaction, such as surfacing, reading, saving, and citing, and aggregates these scores to rank items, allowing for the gauging of credibility and relevance within a user base.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional search engines are used in closed secure environments, then document retrieval is possible, but the ability to surface relevant and important documents is reduced due to smaller user base and continuously streaming content
Solution Approach 1:
The patent changes the parameter of document evaluation from simple citation counts to a multi-level usage score system that weights different interaction types (views, reads, saves, citations) differently. This allows reliable relevance assessment even with smaller user bases by considering the quality and depth of engagement rather than just quantity of interactions
Solution Approach 2:
The patent introduces an intermediary usage scoring mechanism that mediates between user interactions and document ranking. This intermediary layer aggregates and weights various user actions to produce a comprehensive relevance signal, enabling reliable document surfacing despite limited direct user feedback in closed environments
2Measurement precision
If citation tracking is used to assess document quality, then document importance can be measured, but it fails to capture the full spectrum of user engagement levels
Solution Approach 1:
The patent segments document usage into multiple distinct levels (surface views, reads, saves, citations) rather than treating all interactions equally. Each level is assigned a weight reflecting its significance, creating a nuanced quality assessment that captures the full spectrum of user engagement from passive viewing to active citation
Solution Approach 2:
The patent adds dimensional depth to document evaluation by creating a multi-dimensional usage score system. Instead of a single citation count dimension, it incorporates multiple interaction dimensions with different weights, providing a more comprehensive and precise measurement of document quality and relevance
Data Source
AI summary
Provided are techniques for gauging credibility of digital content items. For each digital content item in a list of digital content items, a usage score is identified for each different level of usage and for a user base and the usage score for each different level of usage is added to obtain a total usage score for the digital content item. Each digital content item in the list is ranked using the total score of each digital content item.