Sentiment Aggregation Using K-Valued Model and Confidence Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sentiment aggregation methods struggle to accurately determine aggregate sentiment scores and confidence levels from a corpus of documents, especially when there are ties or near-ties in sentiment scores, and fail to handle linguistic convictions and weak signals effectively.
Innovation Solution
A method that uses a k-valued sentiment model to determine an aggregate sentiment score and confidence level by calculating a raw aggregate sentiment score and selecting the closest k-valued actual score, with confidence determined by the proximity of primary and secondary sentiment scores, and further enhanced by considering data spread and linguistic conviction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sentiment aggregation methods are used to determine aggregate sentiment scores, then the process is simple, but the accuracy and reliability of the aggregate sentiment score deteriorates, especially when there are ties or near-ties in sentiment scores
Solution Approach 1:
The patent changes the parameter space from binary sentiment scores to a k-valued model where k > 2, allowing for multiple sentiment categories (e.g., positive, neutral, negative). This enables more precise aggregation by considering the distribution across multiple sentiment values rather than just two, thereby improving measurement precision while managing complexity through structured parameter transformation.
Solution Approach 2:
The patent introduces a new dimension to sentiment aggregation by incorporating confidence scores alongside sentiment scores. This dimensional expansion allows the system to not only determine the aggregate sentiment but also assess the reliability and strength of that sentiment, addressing the accuracy issue without fundamentally changing the core aggregation mechanism.
2Reliability
If traditional sentiment aggregation methods are used, then the processing is straightforward, but the ability to handle linguistic convictions and weak signals deteriorates
Solution Approach 1:
The patent introduces confidence scores as an intermediary element that mediates between the raw sentiment scores and the final aggregate sentiment determination. This intermediary component allows the system to handle linguistic convictions and weak signals by quantifying the strength and reliability of each sentiment passage before aggregation, thereby improving reliability without requiring complex linguistic analysis mechanisms.
Solution Approach 2:
The patent performs preliminary scoring of individual sentiment passages to generate both sentiment scores and confidence scores before performing the aggregation. This preliminary action allows the system to pre-process and rank sentiment passages based on their strength and reliability, enabling better handling of linguistic nuances during the final aggregation step.
3Measurement precision
If a k-valued model is used to determine aggregate sentiment score, then the precision of sentiment analysis improves, but the complexity of calculating confidence scores increases
Solution Approach 1:
The patent segments the sentiment analysis process into distinct components: sentiment scoring, confidence scoring, and aggregation. By separating the confidence score calculation from the sentiment aggregation, the system can maintain high precision through the k-valued model while managing complexity through modular processing. Each component can be optimized independently.
Solution Approach 2:
The confidence score mechanism serves multiple functions: it quantifies the strength of sentiment passages, handles ties and near-ties, and provides reliability information for the aggregate sentiment. This multi-functionality allows the system to achieve high precision without proportionally increasing complexity, as a single mechanism addresses multiple challenges.
Data Source
AI summary
A mechanism is provided in a data processing system for aggregating sentiment about an entity from a corpus of documents. The mechanism identifies a plurality of sentiment passages in the corpus of documents. Each of the plurality of sentiment passages includes a statement of sentiment about the entity. The mechanism determines a plurality of passage sentiment scores for the plurality of sentiment passages and an actual aggregate sentiment score from the plurality of passage sentiment scores based on a k-valued model. The mechanism determines a sentiment confidence score for the actual aggregate sentiment score based on the raw aggregate sentiment score and the actual aggregate sentiment score and presents the actual aggregate sentiment score and the sentiment confidence score.