Affinity Scoring for Search Relevance via Glossary Probabilities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search engines struggle to provide a comprehensive overview of documents related to a specific topic, as they primarily rely on exact word matches, leading to relevant documents being overlooked.
Innovation Solution
A method that determines the affinity of a piece of content to a particular category by using a glossary defined for that category, which associates words with probability values, and modifies the relevancy of pre-processed content to improve search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If search engines rely on exact word matches, then search results are easy to compute, but relevant documents are overlooked and search coverage is incomplete
Solution Approach 1:
The patent changes the parameter of word matching from binary exact matches to probabilistic affinity scores. Each word is assigned a probability value representing its affinity to a category, allowing the system to compute relevance based on cumulative probability rather than exact matching, thereby improving search result relevance without significantly increasing computational complexity
Solution Approach 2:
The patent introduces an intermediary glossary structure that mediates between raw search queries and document content. The glossary maps words to probability values representing their categorical affinity, serving as a bridge that enables probabilistic matching while maintaining computational efficiency through pre-computed lookup tables
2Adaptability or versatility
If search engines use only words and operators entered by users, then the search system remains simple, but it cannot provide a good overview of documents surrounding a particular topic
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing probability values for words in the glossary before actual search operations. This pre-processing enables the system to quickly determine topic overview by summing pre-computed probability values for words associated with a topic, providing comprehensive topic coverage without adding complexity during the search query processing
Data Source
AI summary
Some embodiments provide a method for determining a relatedness of content items to categories. The method identifies a particular content item, a relevancy score associated with the particular content item, and a set of categories to which the particular content item is classified as related. Based on a set of glossaries associated with the set of categories, the method calculates a set of affinity scores that each represents a degree of relevancy between the particular content item and a category in the set of categories. The method modifies the relevancy score associated with the particular content item based on the calculated set of affinity scores.