Sentiment Analysis Using Intermediate Emotional Score Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing natural language processing systems are inefficient and unreliable in performing sentiment designation without labeled data, requiring time-consuming and labor-intensive labeling processes.
Innovation Solution
A method and apparatus that dynamically generate an overall emotional sentiment score for an input text sequence using an emotional sentiment machine learning model, which processes the input text to create an intermediate emotional sentiment score object and then transforms it into an overall score using a sentiment score transformation object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing natural language processing systems are used for sentiment designation, then the system can process text input, but the computation is inefficient and reliability is poor without labeled data
Solution Approach 1:
The patent segments the sentiment analysis process into multiple intermediate emotional sentiment evaluation dimensions (e.g., joy, sadness, anger, fear) rather than directly computing overall sentiment. Each dimension is evaluated separately by the machine learning model, and results are aggregated to form the final sentiment designation. This segmentation improves both reliability through multiple evaluation perspectives and efficiency by distributing computational load across parallel dimension assessments.
Solution Approach 2:
The patent introduces intermediate emotional sentiment score objects as mediators between the raw text input and the final sentiment designation. These intermediate scores represent evaluations across multiple emotional dimensions and serve as a bridge that transforms unstructured text into structured sentiment information. This intermediary representation layer improves reliability by capturing nuanced emotional content and efficiency by enabling modular processing and reuse of intermediate results.
2Measurement precision
If labeled training data is used to improve sentiment analysis accuracy, then measurement precision improves, but the process becomes time-consuming and labor-intensive
Solution Approach 1:
The patent implements self-service by enabling the system to generate and refine its own training data through automated processes. The machine learning model performs sentiment analysis on unlabeled data, and the results are used to automatically create labeled training examples. This self-service mechanism eliminates the need for manual data labeling while maintaining measurement precision, as the system continuously learns and improves from its own operations.
Solution Approach 2:
The patent changes the parameter of training data requirements by transitioning from a system that requires extensive labeled data to one that can operate effectively with unlabeled or minimally labeled data. This is achieved by modifying the training approach to use unsupervised or self-supervised learning techniques, where the model learns sentiment patterns from the structure and context of the text itself rather than relying on pre-labeled examples. This parameter change dramatically reduces data labeling time while preserving measurement precision.
3Measurement precision
If the system processes the input text sequence as a whole, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the complex task of overall sentiment analysis into multiple simpler intermediate evaluations across different emotional dimensions. Instead of processing the entire text sequence to directly determine overall sentiment, the system first evaluates separate emotional dimensions (joy, sadness, anger, etc.) and then aggregates these intermediate results. This segmentation reduces device complexity by breaking down the processing task into manageable, modular components while maintaining measurement precision through comprehensive multi-dimensional evaluation.
Data Source
AI summary
There is a need for more effective and efficient predictive data analysis solutions and/or more effective and efficient solutions for generating an emotional sentiment score without the use of labelled data. In one example, embodiments comprise receiving an input text sequence, generating an intermediate emotional sentiment score object based at least in part on the input text sequence and by utilizing an emotional sentiment machine learning model, generating an overall emotional sentiment score based at least in part on the intermediate sentiment score object and by utilizing an emotional sentiment score transformation object, and performing one or more prediction-based actions based at least in part on the overall emotional sentiment score.