Composite Search Scoring via Text and Category Blending
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines face challenges in retrieving and scoring Web content effectively due to its unstructured nature, leading to poor quality search results for esoteric content, short-lived content like advertisements, and repetitive content, as they rely primarily on basic text matching that overlooks relevant content and fails to quantify match quality.
Innovation Solution
A system and method that determines a composite score for search results by integrating text match scores with category match scores, using a scoring engine that annotates documents with labels, associates them with categories, and blends text and label match scores to provide a qualitative assessment of content relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If basic text matching is used to search Web content, then the search process is simple and fast, but the quality of search results deteriorates due to literal interpretation and inability to capture qualitative relevance
Solution Approach 1:
The patent segments the search evaluation into multiple independent components: text match score, category match score, and blended score. This allows each component to be calculated separately and then combined, maintaining computational efficiency while improving overall match quality assessment through multi-dimensional evaluation
Solution Approach 2:
The patent creates a composite scoring mechanism that combines text match scores and category match scores into a blended score. This composite approach integrates multiple evaluation criteria (textual relevance and categorical relevance) to produce a more accurate overall assessment of search result quality
2Measurement precision
If categories and labels are integrated into search scoring, then the quality of match assessment improves, but the system complexity increases due to additional processing steps
Solution Approach 1:
The patent performs preliminary actions by pre-associating documents with categories and labels before the actual search query is processed. This preparation work is done in advance, allowing the search system to quickly retrieve and score documents based on pre-computed category associations rather than performing complex categorical analysis during query processing
Solution Approach 2:
The patent introduces category associations as an intermediary layer between the search query and the document collection. Categories act as mediators that bridge the gap between user queries and relevant documents, enabling the system to assess match quality without directly comparing every query term against every document in a complex manner
3Measurement precision
If multiple scoring components are blended, then the accuracy of search result ranking improves, but the computational time increases due to multiple score calculations
Solution Approach 1:
The patent implements partial action by calculating only the necessary scoring components for each search result. Rather than computing all possible metrics for every document, the system calculates text match scores and category match scores selectively, then blends them to produce the final ranking, avoiding unnecessary computational overhead
Data Source
AI summary
A system and method for scoring documents is described. One or more documents are identified responsive to a search criteria. A text match score indicating a quality of match of the identified documents is determined. A category match score is determined over categories. A document-categories score is determined indicating a quality of match between an identified document and a plurality of categories. A search criteria-categories score is determined indicating a quality of match between the search criteria and the categories. An overall score is determined based on the text match score and the category match score.