Sentiment Analysis via Syntactic Constituents and Entity Linking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sentiment analysis systems face challenges in accurately determining sentiment between entities in natural language, particularly in smaller text units like clauses or noun phrases, and fail to capture implicit sentiment due to their rigid frameworks and shallow methods, which limit their flexibility and effectiveness.
Innovation Solution
A computerized method that analyzes natural language to determine sentiment between entities by receiving and processing the language to identify syntactic constituents, linking relevant entities, and calculating an overall sentiment score through sentiment scoring of each constituent, while also generating a confidence score for the assessed sentiment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If fixed frames or templates are used for sentiment analysis, then the analysis structure is simplified, but the system becomes less flexible and less useful for varied text units
Solution Approach 1:
The patent implements a dynamic analysis framework that adapts to different text units (clauses, noun phrases, sentences) rather than using fixed templates. The system dynamically identifies and processes relevant constituents based on the input text structure, enabling flexibility across varied text types while maintaining organized analysis through its modular constituent-based approach.
2Productivity
If shallow methods are used for entity-level sentiment analysis, then the processing is simpler and faster, but implicit sentiment layers are ignored and analysis accuracy is reduced
Solution Approach 1:
The patent segments sentiment analysis into multiple hierarchical layers, processing both explicit and implicit sentiment at different levels (constituent level, phrase level, sentence level). This multi-layered segmentation allows the system to capture implicit sentiment that shallow methods miss, while maintaining processing efficiency through structured modular analysis of each layer.
Solution Approach 2:
The patent adds a dimensional layer to sentiment analysis by introducing implicit sentiment analysis alongside explicit sentiment. This multi-dimensional approach transforms the analysis from a single-layer explicit sentiment detection to a multi-layer framework that simultaneously processes explicit and implicit sentiment dimensions, enhancing accuracy without sacrificing processing speed.
3Device complexity
If atomic sentiment carriers are analyzed in isolation, then the analysis is simpler, but misleading results occur due to contextual operators that modify sentiment
Solution Approach 1:
The patent introduces syntactic constituents as intermediary elements that connect atomic sentiment carriers with their contextual operators. Rather than analyzing sentiment carriers in isolation or treating the entire sentence as a single unit, the system uses constituents (phrases, clauses) as intermediate layers that preserve both the atomic sentiment information and the contextual modification, enabling accurate sentiment classification through this mediating structure.
Data Source
AI summary
A computer implemented method for analyzing natural language to determine a sentiment between two entities discussed in the natural language, comprising the following steps: receiving the natural language at a processing circuitry; analyzing the natural language to determine a syntactic representation which shows syntactic constituents of the analyzed natural language and to determine a sentiment score of each constituent; determining which constituents link the two entities; and calculating an overall sentiment score for the sentiment between the two entities by processing the sentiment score of each constituent of the constituents determined to link the two entities.