Answer-Seeking Query Classification and Passage Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current internet search engines face challenges in identifying and presenting concise answers to answer-seeking queries, as they often return irrelevant or non-specific results due to the lack of effective classification and scoring of query types and answer elements.
Innovation Solution
A search system that classifies queries as answer-seeking by matching terms against question types and answer types, computes scores for text passages based on matching elements, and provides a presentation that prominently displays the highest-scoring answers, using techniques such as n-gram analysis and question/answer type pair generation from training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the search system returns a set of search results for a query, then the user can find relevant resources, but the system cannot provide a concise direct answer to answer-seeking queries
Solution Approach 1:
The patent extracts the essential answer information from search results by identifying characteristic answer elements (such as entity names, numerical values, dates) that directly respond to answer-seeking queries. Instead of presenting all search results, the system extracts and presents only the crucial answer components, thereby reducing information loss while improving retrieval efficiency.
Solution Approach 2:
The patent segments the search result processing into distinct components: query classification to identify answer-seeking queries, answer element identification to extract specific answer components, and selective presentation to display only relevant answer information. This segmentation enables the system to handle different query types differently and provide concise answers where appropriate.
2Measurement precision
If the search system classifies queries and computes scores for passages, then the relevance of answers improves, but the system complexity increases
Solution Approach 1:
The patent implements preliminary action by pre-defining question types with their characteristic answer elements and pre-computing scoring criteria. During query processing, the system simply matches the query against predefined question types and applies predetermined scoring rules, rather than performing complex analysis from scratch. This reduces online processing complexity while maintaining high measurement precision.
Solution Approach 2:
The patent changes parameters by transforming the query processing approach from general-purpose to specialized: it identifies specific question types (who, what, when, where, why, how) and their characteristic answer elements, then applies type-specific scoring criteria. This parameter-based classification system improves answer relevance accuracy while keeping the processing framework manageable through systematic categorization.
3Adaptability or versatility
If the system automatically discovers question and answer types from training data, then the adaptability to different query types improves, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by automatically discovering and storing question types and their characteristic answer elements during an offline training phase using training data. This pre-processing creates a comprehensive question type database that can be quickly referenced during online query processing. The adaptability to different query types is achieved through this pre-built knowledge base, while the time cost is incurred during offline training rather than online processing.
Solution Approach 2:
The patent implements dynamics by enabling the system to adapt to new query types through the automatic discovery mechanism using training data. The question type classification system can learn and incorporate new question patterns and answer elements dynamically. This allows the system to maintain high adaptability to diverse query types while managing processing time through efficient matching against the learned question type database.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating answers to answer-seeking queries. One of the methods includes receiving a query having multiple terms. The query is classified as an answer-seeking query of a particular question type, and one or more answer types associated with the particular question type are obtained. Search results satisfying the query are obtained, and a respective score is computed for each of one or more passages of text occurring in each document identified by the search results, wherein the score for each passage of text is based on how many of the one or more answer types match the passage of text. A presentation that includes information from one or more of the passages of text selected based on the respective score is provided in response to the query.