Contextual Bigram Analysis for Search Query Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional bigram analysis in search engines is computationally intensive and inefficient, especially when dealing with search queries containing multiple terms, as it assumes complete sentences and correct grammar, making it difficult to determine whether consecutive words are intended to form n-grams or remain separate keywords.
Innovation Solution
The implementation of a contextual bigram analysis system that reduces the number of potential bigrams to analyze by identifying and excluding differing search terms from consideration based on context derived from previous search queries, thereby increasing analysis efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional bigram analysis is performed on all potential bigrams in a search query, then comprehensive n-gram identification is achieved, but computational efficiency deteriorates due to the large number of potential bigrams to analyze
Solution Approach 1:
The system performs preliminary analysis by examining previous search queries in the same session to identify terms that are unlikely to form bigrams with the current query terms. This preliminary filtering action is taken before the actual bigram analysis, reducing the number of potential bigrams that need to be analyzed while maintaining identification accuracy.
Solution Approach 2:
The system extracts and excludes certain search terms from bigram analysis based on contextual information from previous queries. By identifying terms that should not be part of bigrams and removing them from the analysis set, the system reduces computational workload without compromising the accuracy of bigram identification for the remaining terms.
2Productivity
If contextual information from previous queries is used to exclude terms from bigram analysis, then processing efficiency is improved, but the complexity of the analysis system increases
Solution Approach 1:
The system segments the bigram analysis process into two distinct phases: (1) a filtering phase that uses previous queries to identify terms to exclude from bigram analysis, and (2) an analysis phase that performs bigram analysis only on the remaining potential bigrams. This segmentation reduces overall system complexity by breaking down the complex task into manageable, independent steps.
3Reliability
If all consecutive word pairs in a search query are analyzed as potential bigrams, then potential n-grams are not missed, but the number of analysis operations increases significantly
Solution Approach 1:
The system performs preliminary filtering using contextual information from previous search queries before conducting the actual bigram analysis. By identifying and excluding terms that are unlikely to form bigrams in advance, the system maintains detection completeness for relevant bigrams while significantly reducing the time required for analysis.
Data Source
AI summary
Methods, systems, and apparatus, including computer program products, in which a first search query in a session is compared to a second search query within the session to determine whether any terms included in the second search query differ from terms in the first search query. If differing search terms are identified in the second search query, the differing search terms can be excluded from bigram analysis with other search terms included in the second query.