Question Answering System Lat Normalization Infographic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional question answering systems face challenges in aggregating and displaying answers to natural language questions, particularly when users pose verbose or multi-pronged queries, and struggle to present data in a visually consumable format, as users may not be aware of the infographic containing relevant data points.
Innovation Solution
The system clusters questions by contextual theme, normalizes lexical answer types within a question set, and aggregates answers across multiple sessions, transforming discordant questions to a base lexical answer type for graphical representation in an infographic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the system aggregates answers from multiple question sets with different lexical answer types, then the comprehensiveness of information retrieval is improved, but the complexity of answer normalization and aggregation increases
Solution Approach 1:
The system transforms discordant lexical answer types into a base LAT by changing the parameter of answer type classification. This allows diverse answers (e.g., dates, times, durations) to be normalized into a unified format, enabling aggregation across different question sets while maintaining information comprehensiveness
Solution Approach 2:
The base lexical answer type acts as an intermediary category that mediates between diverse answer types. By introducing this intermediate normalization layer, the system can aggregate answers from questions with different LATs without directly complex interactions between all possible answer type combinations
2Stability of the object's composition
If the system transforms discordant questions to a base lexical answer type, then the consistency of answer format is improved, but the processing time for question transformation increases
Solution Approach 1:
The system performs preliminary classification of questions into question sets based on their lexical answer types before aggregation. By pre-organizing questions and identifying base LATs in advance, the actual transformation and aggregation process during query execution is accelerated, reducing the perceived processing time
3Measurement precision
If the system clusters questions by contextual theme, then the relevance of aggregated answers is improved, but the complexity of contextual analysis increases
Solution Approach 1:
The system segments the overall question set into smaller contextual themes or clusters before performing aggregation. This segmentation approach allows contextual analysis to be applied to smaller, more manageable groups, reducing the complexity of analyzing all questions simultaneously while maintaining answer relevance through theme-based organization
Data Source
AI summary
A technique for transforming questions for a question answering (QA) system includes clustering received questions by contextual accordance into a question set. A base lexical answer type (LAT) for the question set is determined. LATs of discordant questions in the question set are transformed to the base LAT. Answers for the questions in the question set are aggregated. The aggregated answers are then displayed.