Deep Question Answer Search System for Direct Answer Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engine technologies cannot directly provide answers to user queries, relying on keyword matching and document summarization, leading to irrelevant and poorly presented search results, which negatively impact user experience.
Innovation Solution
An answer searching method based on deep question and answer technology, utilizing a combination of webpage analysis, deep learning, and template matching methods to identify and display final answers directly in search results, improving relevance and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional information retrieval technology and document summarization technology are used, then webpages with higher relevancy can be provided as search results, but users cannot directly obtain answers and must manually extract required information, leading to poor search result display effect
Solution Approach 1:
The patent introduces an answer extraction system as an intermediary between the search engine and user. This system automatically extracts key information from search results and presents it in a structured answer format, mediating the gap between raw search results and user needs without requiring manual information extraction
Solution Approach 2:
The system enables self-service by automatically performing the information extraction task that would otherwise require user effort. The answer extraction mechanism autonomously processes search results, identifies relevant information, and formats answers, allowing users to obtain answers directly without manual intervention
2Measurement precision
If multiple answer processing methods (webpage analysis, deep learning, template matching) are combined, then answer accuracy and relevance are improved, but system complexity increases
Solution Approach 1:
The patent segments the answer extraction process into three distinct methodological components: webpage analysis for structural information extraction, deep learning for semantic understanding, and template matching for standardized answer patterns. Each segment handles specific aspects of the extraction task, allowing the system to manage complexity through modular division while achieving high accuracy through the combination of specialized approaches
Solution Approach 2:
The system achieves multi-functionality by designing a unified answer extraction framework that can handle diverse query types and answer formats through multiple processing methods. The same system architecture supports different extraction approaches (webpage analysis, deep learning, template matching) depending on the query characteristics, making the system versatile without requiring separate systems for each function
Data Source
AI summary
The present disclosure provides an answer searching method and device based on deep question and answer. The method includes: receiving a query statement; obtaining webpages containing candidate answers by searching according to the query statement; obtaining a first candidate answer, a second candidate answer and a third candidate answer based on a webpage analysis method, a deep learning method and a template matching method respectively; determining a final answer according to the first candidate answer, the second candidate answer and the third candidate answer; and displaying the final answer.