Sentiment Lexicon Expansion via Syntactic Dependency Parsing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The manual and labor-intensive process of sentiment lexicon acquisition and expansion is inefficient, particularly due to domain sensitivity issues, leading to poor classification accuracy when using generic sentiment lexicons across different domains.
Innovation Solution
A semi-supervised method for automating sentiment lexicon expansion using a domain-specific corpus and generic sentiment lexicon, applying syntactic dependency rules to identify and filter sentiment candidates, and evaluating their polarity, which are then added to the lexicon, enhancing domain adaptation and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a generic sentiment lexicon is used for domain-specific text classification, then the lexicon coverage is broad, but the classification accuracy deteriorates due to domain sensitivity
Solution Approach 1:
The patent segments the sentiment lexicon into domain-specific subsets by parsing domain-specific corpora into dependency relations and identifying sentiment candidates within each domain context. This allows the lexicon to be divided into targeted domain segments rather than using a single generic lexicon for all domains.
Solution Approach 2:
The patent applies local quality by adapting the sentiment lexicon to specific domain characteristics through syntactic dependency rules tailored to each domain. Each domain receives a customized lexicon version with sentiment candidates selected based on domain-specific linguistic patterns and contexts.
2Measurement precision
If manual processes are used for sentiment lexicon acquisition and expansion, then the lexicon quality can be controlled, but the operational cost and time consumption increase significantly
Solution Approach 1:
The patent implements self-service by enabling the sentiment lexicon to automatically expand through syntactic dependency rules that identify sentiment candidates from domain-specific corpora without requiring manual intervention for each candidate evaluation. The system serves itself by generating and filtering sentiment candidates autonomously.
Solution Approach 2:
The patent replaces the manual mechanical process of lexicon acquisition with an automated computational system using syntactic dependency parsing and rule-based identification. This substitutes human labor with algorithmic processing while maintaining quality control through structured evaluation criteria.
3Measurement precision
If manual sentiment lexicon expansion is performed to improve domain adaptation, then the classification accuracy improves, but the labor intensity and time requirements increase
Solution Approach 1:
The patent applies preliminary action by pre-defining syntactic dependency rules and thresholds for sentiment candidate identification before actual lexicon expansion begins. This preparation work is done once and then reused across multiple domain adaptations, reducing the time required for each subsequent expansion task.
Solution Approach 2:
The patent replaces manual lexicon expansion with automated syntactic dependency parsing and rule-based sentiment candidate identification. This computational approach processes domain corpora much faster than manual methods while maintaining systematic quality control through predefined evaluation criteria.
Data Source
AI summary
Systems and methods for sentiment lexicon expansion receive at least a domain specific corpus comprising a plurality of words, and a generic sentiment lexicon; parse the plurality of words in the domain specific corpus into a plurality of dependency relations; identify, using one or more syntactic dependency rules and at least one of the plurality of dependency relations, a set of one or more sentiment candidates in the domain specific corpus; filter from the set of one or more sentiment candidates any sentiment candidate having an expected performance below a predefined threshold; sample the filtered set of one or more sentiment candidates to be used in a qualitative evaluation; and, for each sentiment candidate that passes the qualitative evaluation, add the sentiment candidate to the generic sentiment lexicon.