Dynamic Sentiment Dictionary via Conjunction Analysis

Solution Overview

Problem

Existing sentiment analysis tools face challenges in accurately analyzing social media data due to the informal nature of social media writing, which is not well-covered by current sentiment dictionaries, and the difficulty in updating these dictionaries to include new words, leading to low recall and accuracy.

Innovation Solution

The method involves identifying seed words, expanding them using synonyms and antonyms, and using a constrained symmetric nonnegative matrix factorization (CSNMF) algorithm to generate sentiment dictionaries by combining social media data with lexical databases like WordNet, calculating sentiment scores based on conjunction relationships, and outputting polarity strength scores for each word.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If a lexicon-based approach with existing sentiment dictionaries is used, then the sentiment analysis can be performed without requiring large amounts of human-labeled training data, but the analysis accuracy is low because the dictionaries do not cover informal and spoken words used in social media

Engineering Contradiction:

Improveease of implementing sentiment analysisVSAvoidsentiment analysis accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The sentiment dictionary is made dynamic and updateable. The system allows frequent updates to include newly generated words and informal language used in social media, transforming the static nature of traditional sentiment dictionaries into a flexible, evolving resource that adapts to changing language patterns.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the parameters of sentiment analysis by using adjective seed words and conjunction relationships (particularly 'but') to generate polarity scores. This approach transforms the traditional lexicon-based method by incorporating statistical frequency analysis and contextual relationships, thereby improving accuracy while maintaining ease of implementation.

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If existing sentiment dictionaries are used, then the implementation is simple, but the recall is low because the dictionaries are not able to update frequently to include newly generated words

Engineering Contradiction:

Improveease of using sentiment dictionaryVSAvoidrecall of sentiment analysis

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The sentiment dictionary is made dynamic and updateable. The system allows frequent updates to include newly generated words and informal language used in social media, transforming the static nature of traditional sentiment dictionaries into a flexible, evolving resource that adapts to changing language patterns.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system enables automatic updating of the sentiment dictionary using adjective seed words and conjunction relationships from social media data. This self-service mechanism allows the dictionary to automatically incorporate new words and language patterns without requiring manual curation, thereby improving recall while maintaining ease of operation.

Inventive Principle:

Principle #25Self-service

3Measurement precision

If adjective seed words and conjunction relationships are used to build sentiment dictionaries, then the precision and recall of sentiment analysis are improved, but the complexity of the system increases

Engineering Contradiction:

Improvesentiment analysis accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system changes the parameters of sentiment analysis by using adjective seed words and conjunction relationships (particularly 'but') to generate polarity scores. This approach transforms the traditional lexicon-based method by incorporating statistical frequency analysis and contextual relationships, thereby improving accuracy while maintaining ease of implementation.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system uses conjunction relationships (especially 'but') as intermediaries to infer sentiment polarity. By leveraging these linguistic connectors that appear in social media data, the system can automatically determine sentiment directions without requiring complex machine learning models, thus improving accuracy while controlling system complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS8725495B2Systems, methods and devices for generating an adjective sentiment dictionary for social media sentiment analysis

Publication Date: 2014.05.13 XEROX CORP

AI summary

Embodiments generally relate to systems and methods for generating a sentiment dictionary and calculating sentiment scores of adjectives within the sentiment dictionary. A set of seed words can be identified and expanded using synonyms and antonyms of the set of seed words. Social media data can be parse to identify adjectives that link to the set of seed words with the words “and” or “but.” Matrices representing the attraction and repulsion among the linked adjectives can be generated. A factorization algorithm can be minimized to determine an output matrix that comprises positive and negative sentiment scores for each of the adjectives. In embodiments, a sentiment score for part of all of the social media data can be calculated using the output matrix, and one or more parts of the social media data can be classified as a positive or negative sentiment.