Search Result Ranking via Natural Language Classification Confidence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search systems fail to accurately provide search results based on the intention behind a search query, leading to inefficient use of computing resources and inaccurate results due to the lack of consideration for the subject or intent of the query.
Innovation Solution
Implementing natural language classification (NLC) techniques to determine the confidence levels of search query intentions and using these confidence levels to filter and rank search results, ensuring that the ratio of search results matches the query's intended subject or intent.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If natural language classification is applied to determine search query intention, then search result accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces a natural language classification server as an intermediary component between the search system and the query. This server receives search queries, applies NLC techniques to determine intent and confidence levels, and returns classification results to the search system. By externalizing the NLC functionality to a separate server, the patent adds the capability to improve search accuracy while isolating the complexity in a dedicated component rather than embedding it throughout the entire search system.
Solution Approach 2:
The patent segments the search system into distinct functional components: the search system itself, the natural language classification server, and the result filtering mechanism. Each component performs a specific function - the search system processes queries, the NLC server determines intent with confidence levels, and the filtering mechanism applies these levels to rank results. This segmentation allows the system to gain the benefits of NLC while managing complexity through modular architecture.
2Reliability
If search results are filtered based on confidence levels, then relevant information is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by determining the confidence levels and classification of search query intentions before generating and filtering search results. The natural language classification server analyzes the query intent and assigns confidence levels in advance, so that when search results are generated, they can be immediately filtered and ranked according to these pre-determined confidence levels. This preliminary classification reduces the processing time during result generation compared to analyzing each result individually.
Data Source
AI summary
A computer-implemented method includes: receiving, by a computing device, a search query from a client device; obtaining, by the computing device, classification and confidence information by applying natural language classification to the search query; generating, by the computing device, search results based on the classification and confidence information, wherein a ratio of the search results is based on the classification and confidence information; and providing, by the computing device, the search results to the client device.