Sentiment Polarity Vector Construction for NLP Phrase Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current natural language processing methods struggle to accurately determine the sentiment polarity of new sentiment phrases, especially in languages like Chinese and Japanese, where phrases can be formed by combining multiple words, leading to a need for improved methods to analyze and update sentiment dictionaries efficiently.
Innovation Solution
A natural language processing method that determines the sentiment polarity of a phrase by calculating polarity characteristic vectors based on the polarity characteristic information of its constituent words, using a classifier to analyze and output the sentiment polarity, and automatically extends the sentiment dictionary by adding newly analyzed phrases, allowing for more accurate and user-configurable sentiment analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a sentiment dictionary is built using traditional methods, then existing sentiment phrases can be analyzed, but new sentiment phrases cannot be accurately determined
Solution Approach 1:
The phrase is segmented into constituent words, and polarity characteristic information is extracted from each word individually. This segmentation allows the system to analyze new phrases by breaking them down into known word components, enabling accurate sentiment determination for previously unseen phrases.
Solution Approach 2:
The patent introduces a new dimension of analysis by creating polarity characteristic vectors that combine polarity characteristic values from multiple words. This vector-based approach transforms the problem from dictionary lookup to mathematical computation, enabling the system to handle new sentiment phrases through vector operations rather than requiring pre-existing dictionary entries.
2Device complexity
If sentiment polarity is determined using word-level analysis, then processing is simplified, but phrase-level context is lost
Solution Approach 1:
The patent merges the polarity characteristic values from multiple individual words into a unified polarity characteristic vector for the entire phrase. This combining process preserves phrase-level context by considering the collective contribution of all constituent words, while maintaining computational simplicity through vector operations.
3Adaptability or versatility
If a comprehensive sentiment dictionary is manually created, then coverage is improved, but maintenance becomes difficult as new phrases emerge
Solution Approach 1:
The system performs self-service by automatically determining the sentiment polarity of new phrases through computational methods rather than requiring manual dictionary updates. The polarity characteristic vectors are computed on-demand from constituent word information, enabling the system to adapt to new sentiment phrases without human intervention.
Data Source
AI summary
A natural language processing method includes obtaining a to-be-processed phrase, where the to-be-processed phrase includes M words, determining polarity characteristic information of m to-be-processed words in the M words, where polarity characteristic information of an ith word in the m to-be-processed words includes n polarity characteristic values, and each polarity characteristic value corresponds to one sentiment polarity, determining a polarity characteristic vector of the to-be-processed phrase based on the polarity characteristic information of the m to-be-processed words, where the polarity characteristic vector includes n groups of components in a one-to-one correspondence with n sentiment polarities, and determining a sentiment polarity of the to-be-processed phrase based on the polarity characteristic vector of the to-be-processed phrase using a preset classifier, and outputting the sentiment polarity.