Question Answering System User Expectation Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current question answering systems are inefficient in processing user input in natural language due to noise and ambiguity, as they often fail to accurately identify the key points or user expectations within questions, leading to less effective relevance measurement and answer retrieval.
Innovation Solution
An automated support system leveraging machine learning algorithms, specifically supervised and unsupervised learning techniques, to extract user expectations by distinguishing between context and key points in user-submitted questions, utilizing a binary classification model to improve answer retrieval and knowledge sharing in online social networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current question answering systems process user input as is, then the system operation is simple, but the answer retrieval accuracy deteriorates due to noise and ambiguity in user questions
Solution Approach 1:
The system segments user questions into context portions and key point portions using machine learning classification. This segmentation allows the system to focus on the key points for answer retrieval while treating context as background information, thereby improving answer retrieval accuracy without requiring complete reprocessing of the entire question.
Solution Approach 2:
The patent introduces an expectation extractor model as an intermediary component between the user question and the answer retrieval system. This model acts as a mediator that extracts and identifies user expectations, transforming the noisy input question into a structured representation that improves downstream retrieval accuracy.
2Ease of operation
If users express their queries in natural language, then the ease of operation is improved, but the measurement precision of user intent deteriorates due to noise and ambiguous descriptions
Solution Approach 1:
The system enables users to express queries in natural language without requiring them to manually structure or clean their input. The machine learning model automatically performs the extraction and identification of key points and user expectations, making the system self-serve in correcting the noise and ambiguity that users introduce through natural language expression.
Solution Approach 2:
The patent changes the parameter representation of user questions by transforming them from raw natural language text into classified portions (context vs. key points) and extracted user expectations. This parameter transformation maintains the ease of natural language input while improving the precision of user intent representation for retrieval purposes.
3Measurement precision
If the system asks for user clarification when ambiguity exists, then the measurement precision improves, but the productivity deteriorates due to additional interaction steps
Solution Approach 1:
The system performs preliminary action by automatically extracting and identifying user expectations before the retrieval process begins. The expectation extractor model proactively clarifies ambiguous intent by identifying key points and user expectations from the input question, eliminating the need for subsequent clarification interactions with the user.
Solution Approach 2:
The patent replaces the mechanical interaction process of asking users for clarification with an automated machine learning-based expectation extraction system. This substitution maintains measurement precision by accurately identifying user intent while preserving productivity by avoiding additional interaction steps.
Data Source
AI summary
Method and system for identifying user expectations in question answering in an on-line social network system are described. The automated support system is configured to address the technical problem of optimization of the processing of user input submitted to a computer in the form of a natural language. The automated support system uses machine learning algorithms to automatically extract, from the user input, information indicative of the user's expectations and obtain data relevant to the input based on said information indicative of the user's expectations.