Automated Text Analysis for User Feedback Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual evaluation of user feedback, which often includes diverse types of information, is time-consuming and inefficient for administrators, as it requires sorting through relevant suggestions for software or hardware improvements amidst irrelevant data.
Innovation Solution
An automated system that analyzes free-form text using an ontology-based approach to identify user suggestions by tagging text segments with classes and recognizing patterns, thereby extracting actionable insights and providing them to administrators for evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual evaluation of user feedback is performed, then administrators can identify bugs and issues, but the process is time-consuming and inefficient
Solution Approach 1:
The patent replaces the manual mechanical evaluation process with an automated text analysis system that uses natural language processing, machine learning, and pattern recognition algorithms to identify suggestions, bugs, and issues in user feedback, thereby eliminating time-consuming manual reading and analysis while maintaining identification accuracy
Solution Approach 2:
The system creates structured representations (copies) of unstructured user feedback text by mapping text segments to ontology classes and generating standardized suggestion formats, enabling automated processing and analysis without requiring administrators to read the original free-form text
2Loss of information
If administrators manually analyze diverse feedback information, then they can identify relevant suggestions, but they must sort through large volumes of irrelevant data
Solution Approach 1:
The patent extracts only the relevant information from diverse user feedback by mapping text segments to specific ontology classes (such as suggestion, bug, feature request) and filtering out irrelevant content, thereby isolating actionable insights from large volumes of mixed feedback data without losing important suggestions
Solution Approach 2:
The system segments user feedback into distinct text segments and maps each segment to specific ontology classes, allowing individual analysis and classification of different types of information (suggestions, bugs, questions, irrelevant content) separately, which improves both extraction completeness and processing efficiency
3Productivity
If automated text analysis is implemented, then evaluation time is reduced, but the system complexity increases
Solution Approach 1:
The patent implements a universal ontology-based framework that can handle multiple types of feedback (suggestions, bugs, feature requests, questions) and map them to standardized classes, allowing a single automated system to perform diverse analysis functions without requiring separate specialized systems for each feedback type
Solution Approach 2:
The system introduces an ontology as an intermediary layer between raw user feedback text and the final analysis results, providing a standardized intermediate representation that simplifies the complexity of direct text analysis while enabling automated processing and maintaining system manageability
Data Source
AI summary
Free-form text in a document can be analyzed using natural-language processing to determine actionable items specified by users in the text or to provide recommendations, e.g., by automatically analyzing texts from multiple users. Words or phrases of the text can be mapped to classes of a model. An actionable item can be determined using the mapped words or phrases that match a selected grammar pattern. Items can be ranked, e.g., based on frequency across multiple documents. In some examples, the classes can include a suggestion-indicator class or a modal-indicator class, and the selected grammar pattern can include one of those classes. In some examples, the mapping can use a dictionary. A new term not in the dictionary can be automatically associated with classes based on attributes of the new term and of terms in the dictionary, e.g., the new term's part of speech or neighboring terms.