Sentence-Level Sentiment Analysis for Noise-Reduced Comment Mining
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sentiment analysis technologies suffer from decreased accuracy due to the inclusion of expressions unsuitable for analysis within the entire posted document, leading to noise and reduced precision.
Innovation Solution
A sentiment analysis system that identifies and analyzes only specific sentences within a comment rather than the entire document, using clustering and morphological analysis to extract aspect terms, opinion words, and polarity, thereby increasing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the entire posted document is used as the target of sentiment analysis, then the analysis covers all content, but the accuracy decreases due to noise from unsuitable expressions
Solution Approach 1:
The patent segments the entire posted document into individual sentences and performs sentiment analysis on each sentence separately. This allows the system to identify and exclude sentences containing expressions unsuitable for sentiment analysis (such as emojis, images, or non-text content), thereby improving accuracy while processing only the relevant portion of the document.
Solution Approach 2:
The patent extracts and removes sentences that contain expressions unsuitable for sentiment analysis from the entire posted document. By filtering out these problematic sentences before analysis, the system eliminates noise that would otherwise degrade the accuracy of sentiment analysis results.
2Measurement precision
If clustering is applied to identify sentiment-related sentences, then analysis accuracy improves, but processing complexity increases
Solution Approach 1:
The patent performs clustering analysis as a preliminary step to identify sentences containing sentiment expressions before conducting the actual sentiment analysis. This preliminary classification organizes the data structure and identifies target sentences in advance, making the subsequent sentiment analysis more efficient and accurate despite the added processing step.
Data Source
AI summary
Provided is a sentiment analysis system including at least one processor configured to: acquire a first comment which relates to a service, and which is input by a user who uses the service; execute clustering relating to the first comment; acquire a sentiment word relating to a sentiment about the service based on an execution result of the clustering; and analyze the sentiment in a first sentence included in the first comment based on the sentiment word.