Feature-Level Review Analysis via Polarity Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing review systems rely on overall product ratings, which can be misleading as they do not account for specific features relevant to individual users, leading to irrelevant feedback and increased time consumption for users comparing products.
Innovation Solution
A method and system that analyze user-generated content to calculate feature scores by determining the polarity of mentions for each product feature, allowing users to view scores for specific features and filter reviews based on relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If overall product ratings are used, then review analysis is simple, but the information is misleading and not relevant to individual users
Solution Approach 1:
The patent segments the overall product rating into individual feature ratings. Each feature (e.g., battery life, screen quality, performance) is analyzed separately with its own polarity score, allowing users to see detailed breakdowns rather than a single aggregated rating. This segmentation preserves feature-specific information while maintaining analytical simplicity through automated processing.
Solution Approach 2:
The patent adds a new dimension to review analysis by introducing feature-level polarity scores alongside or instead of overall ratings. This transforms the single-dimensional overall rating into a multi-dimensional assessment across multiple product features, enabling users to evaluate products based on specific attributes relevant to their needs.
2Loss of information
If detailed feature analysis is provided, then relevant information is improved, but time consumption for users increases
Solution Approach 1:
The system performs automated feature extraction and polarity analysis without requiring user intervention. The patent uses natural language processing to automatically identify features mentioned in reviews and determine their polarity, generating feature-level scores autonomously. This self-service approach provides detailed feature analysis while eliminating the time users would otherwise spend manually analyzing reviews.
Solution Approach 2:
The patent replaces the mechanical process of manual review reading and analysis with automated computational methods. Natural language processing algorithms and sentiment analysis systems automatically extract features and determine polarity, substituting human cognitive effort with machine-based text analysis that processes reviews rapidly and consistently.
3Loss of information
If all reviews are displayed, then complete information is provided, but relevant information is lost due to clutter
Solution Approach 1:
The patent extracts and highlights only the feature-specific information relevant to user needs from the complete set of reviews. By identifying and separating feature-related mentions and their polarities, the system presents extracted relevant information (feature scores and associated review excerpts) while filtering out unrelated content, making information retrieval easy without losing complete review data.
Solution Approach 2:
The patent applies different presentation qualities to different parts of the review data based on relevance. Feature mentions with high polarity scores or strong sentiment are highlighted or emphasized, while neutral or irrelevant portions are minimized or omitted from prominent display. This local quality differentiation helps users quickly identify important information without being overwhelmed by complete review text.
Data Source
AI summary
A system and method for analyzing reviews is disclosed herein. User-generated content (UGC) such as on-line reviews of products can be broken up in to different features of the products being reviewed. The UGC is analyzed to find each mention of each feature. Then a tag cloud or other visual indicia of the features is created. The tag cloud or other visual indicia displays a certain subset of the features, with an indication of how often certain features are discussed. The indication of how often features are discussed can be the font size of the tag cloud or other visual indicia. Other embodiments are also disclosed.