Sentiment Analysis System Using Phase Transition Keyword Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face difficulties in accurately identifying and analyzing sentiment expressed in documents across various sources on the internet, such as expert and casual user opinions, making it challenging to gauge opinions on specific subjects like hotel quality.
Innovation Solution
A sentiment analysis system comprising a sentiment server with modules for crawling, keyword extraction, filtering, and classification, which identifies sentiment signatures within a predetermined distance of keywords, generates lists of keywords and categories, and determines sentiment based on these signatures, using phase transition formulas and machine learning techniques like support vector machines and neural networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional search engines are used to identify documents, then users can find information on subjects of interest, but users cannot identify sentiment expressed by users on particular subjects
Solution Approach 1:
The patent segments the document analysis process into distinct modules: crawling module for data collection, keyword extraction module for identifying key terms, filtering module for removing irrelevant content, and classification module for sentiment analysis. This segmentation allows each module to specialize in specific tasks, improving overall sentiment detection capability while maintaining manageable system complexity
Solution Approach 2:
The patent introduces intermediary components including a sentiment lexicon that serves as a bridge between raw text and sentiment classification, and intermediate processing layers that transform unstructured document data into structured sentiment information. These intermediaries enable the system to capture and analyze sentiment expressions that traditional search engines miss
2Quantity of substance
If comprehensive document crawling is performed to analyze user opinions, then more sentiment data can be collected, but system complexity and processing requirements increase
Solution Approach 1:
The patent applies preliminary action by pre-processing documents during crawling to extract and store key features, keywords, and initial sentiment indicators. This preliminary processing reduces the complexity of subsequent analysis stages while maintaining comprehensive data collection, as the heavy lifting of data preparation is completed upfront
Solution Approach 2:
The patent implements local quality by applying different processing strategies to different parts of the system: the crawling module uses broad data collection strategies, while the classification module applies targeted sentiment analysis techniques. This allows the system to handle large volumes of data with appropriate complexity only where needed, rather than uniformly across the entire system
Data Source
AI summary
A system, computer readable storage medium storing instructions, and computer-implemented method for determining sentiment expressed in documents is disclosed. A document is received from a plurality of documents. A sentence in the document that includes at least one sentiment signature within a predetermined distance of at least one keyword from a list of keywords is identified, wherein the list of keywords is extracted from the plurality of documents and is filtered using a phase transition formula, and wherein the at least one sentiment signature corresponds to an expression of at least one sentiment in the sentence. At least one category corresponding to the at least one keyword of the sentence is determined, wherein the at least one category is included in a list of categories that is generated using the list of keywords. At least one sentiment corresponding to the at least one category is determined based on the at least one sentiment signature.