Topic Label Sentiment Polarity Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current topic modeling techniques often misrepresent sentiment by generating topic labels that include one sentiment despite containing contradictory sentiments, and they lack the ability to generate dynamic, sentiment-agnostic or sentiment-neutral labels effectively, leading to the removal of potentially relevant information.
Innovation Solution
A computer-implemented method that generates both sentiment-oriented and sentiment-neutral topic labels by calculating TF-IDF values, receives a selected sentiment polarity from a user, and identifies or removes topic labels and documents based on this preference, ensuring accurate and relevant information is provided.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If sentiment analysis is integrated into topic modeling to generate sentiment-oriented topic labels, then the system can provide sentiment information, but it misrepresents sentiment by generating labels with contradictory sentiments
Solution Approach 1:
The patent segments topic labels into sentiment-oriented and sentiment-neutral categories, allowing the system to provide sentiment information when needed while maintaining accuracy by selecting only consistent sentiment labels for each topic
Solution Approach 2:
The system dynamically adjusts topic label selection based on sentiment polarity parameters, enabling flexible control over whether sentiment information is included in the output while maintaining reliability through consistency checks
2Ease of operation
If sentiment-agnostic topic labels are generated by removing topic labels with sentiment, then sentiment-neutral labels are provided, but potentially relevant information is lost
Solution Approach 1:
The system dynamically adapts to user preferences by accepting sentiment polarity parameters that determine whether sentiment-oriented or sentiment-neutral labels are returned, allowing users to choose their preferred output type without information loss
Solution Approach 2:
The patent changes the sentiment polarity parameter to control the output format, enabling the same topic modeling system to produce either sentiment-oriented or sentiment-neutral labels based on user needs without permanently removing any information
Data Source
AI summary
Described are techniques for topic modeling including a computer-implemented method of generating a plurality of topic labels corresponding to a plurality of documents clustered into a plurality of topics, where the plurality of topic labels include a sentiment-oriented topic label and a sentiment-neutral topic label. The method further comprises calculating term frequency-inverse document frequency (TF-IDF) values for respective topic labels and corresponding pluralities of documents. The method further comprises receiving a selected sentiment polarity from a user device. The method further comprises identifying a subset of the plurality of topic labels that satisfy the selected sentiment polarity. The method further comprises transmitting at least one topic label of the subset of the plurality of topic labels to the user device, where the at least one topic label has a higher TF-IDF value than other topic labels in the subset of the plurality of topic labels.