Search System Dominant Object Detection via N-gram Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search systems fail to automatically identify the dominant object in complex natural language-based search queries, leading to inconsistent search results that do not align with user intentions.
Innovation Solution
The method involves segmenting search queries into n-grams, extracting and resolving implicit and explicit aspects, and using knowledge graphs to determine the dominant object by comparing ranked categories, which enables enhanced search services such as modified search constraints and alternative results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional search systems process complex natural language queries, then they can handle diverse user inputs, but they fail to accurately identify the dominant object leading to inconsistent results
Solution Approach 1:
The query is segmented into a plurality of n-grams, where each n-gram includes an ordered sequence of one or more tokens from the query. This segmentation allows the system to analyze different parts of the query separately and identify which segment represents the dominant object the user is searching for.
Solution Approach 2:
Knowledge graphs are introduced as an intermediary to compare ranked categories of candidate n-grams against the overall query. The knowledge graph acts as a mediator that leverages pre-existing structured knowledge to determine which n-gram represents the dominant object, resolving the ambiguity in complex natural language queries.
2Measurement precision
If search systems return results that specifically match the search query, then precision is improved, but results may be sparse or unavailable when the query is complex
Solution Approach 1:
The search system dynamically pivots the search operation based on the detected dominant object. When the dominant object is identified, the system can adaptively adjust the search strategy - either maintaining strict matching for that object or expanding to related concepts if results are sparse, thus balancing precision and result quantity dynamically.
Solution Approach 2:
The system changes search parameters based on the detected dominant object. By identifying the core object through n-gram analysis and knowledge graph comparison, the system can modify search constraints and parameters to return both precise matches and relevant alternative results, adjusting the balance between precision and quantity based on the dominant object's context.
3Measurement precision
If the search system segments queries into n-grams and uses knowledge graphs, then dominant object detection is improved, but system complexity increases
Solution Approach 1:
Knowledge graphs are pre-computed and structured before the search operation. This preliminary action allows the system to leverage existing structured knowledge during query processing without performing complex computations in real-time, thus improving dominant object detection accuracy while managing system complexity through pre-processing.
Solution Approach 2:
The query processing is segmented into distinct steps: n-gram segmentation, aspect extraction, candidate identification, and knowledge graph comparison. This segmentation of the processing pipeline makes the complex system more manageable and allows each component to be optimized independently, reducing overall system complexity while maintaining high detection accuracy.
Data Source
AI summary
Various methods and systems for determining a dominant object of a query and employing the dominant object to provide enhanced search services are discussed. A query is segmented into a set of n_grams. Entity extraction and resolution (EER) methods are employed to determine implicit and explicit aspects for each n_gram. N_grams that include explicit aspects are pruned from the set of n_grams and a pruned set of candidate n_grams is generated from the non-pruned n_grams. Knowledge graphs are employed to generate a ranked list of associated categories for each candidate n_gram. A ranked list of categories associated with the un-segmented query is generated based on knowledge graphs. The candidate n_gram with the highest ranked associated category that is also a highly ranked category associated with the un-segmented query is selected as the dominant object of the query. Enhanced search results are provided based on the determined dominant object.