Emotion analysis and prediction system based on automatic data collection

Abstract

The invention discloses an emotion analysis and prediction system based on automatic data collection. The system comprises a user interaction module, a data management module, an emotion analysis module, and an emotion prediction module, wherein the user interaction module receives an emotion analysis instruction input by a user, and provides a visible interface for the user to check information;the data management module collects and manages data to undergo emotion analysis, crawls and stores historical text information or real-time text streams under preset conditions through three crawlerslike a webpage, an API and a URL; the emotion analysis module performs emotion analysis on the data to undergo emotion analysis, and delivers a comprehensive emotion analysis result by combining different emotion analysis processes, wherein emotion analysis comprises word property marking, syntactic structure analysis and emotion orientation analysis; and the emotion prediction module predicts emotion orientation corresponding to the data to undergo emotion analysis according to a comprehensive emotion analysis result in a time period. The system can collect data automatically and deliver emotion prediction results based on historical data in the process.

Description

technical field [0001] The present invention relates to the technical field of emotion analysis and emotion prediction, and more specifically, relates to an emotion analysis and prediction system based on automatic data collection. Background technique [0002] With the rapid development of social platforms (Twitter, Facebook, etc.), e-commerce platforms (Amazon, Taobao, etc.) and review sites (yelp, Douban, etc.) and other information sharing platforms. Various unstructured text content in the network is also growing explosively, so there is an urgent need to analyze and judge the emotional tendency expressed in the text through Natural Language Processing (Natural Language Processing, NLP) technology, such as judging the positive and negative emotions of the text Or analyze emotional states such as sadness, happiness, and anger that occur in text. On this basis, the research of sentiment analysis (also known as comment mining or opinion mining) came into being. For examp...

AI Technical Summary

Problems solved by technology

[0008] In view of the defects of the prior art, the purpose of the present invention is to solve the problem that the analysis results of different existing sentiment analysis methods cannot be visually compared, lack a unified text resource management platform, consume a lot of time, and have a single data source and the crawled data cannot be as needed. Customization, unable to adjust technical issues such as sentiment analysis model based on user feedback

Images