Weibo Public Opinion Analysis System

English document | 中文文档

This project is a Social Network Public Opinion Analysis System designed for monitoring, analyzing, and predicting public opinion trends using data from social media platforms such as Weibo.

Keywords: Deep Learning, Web Scraping, Full-Stack Development, Natural Language Processing (NLP), Transformers, Flask, Sentiment Analysis, Topic Classification, Data Visualization, Real-time Monitoring, Machine Learning

Features

• Real-time Data Collection: Scrapes and processes data from social platforms.
• Data Cleaning & Processing: Cleans and processes collected data for analysis.
• Topic Classification: Categorizes posts and comments into relevant topics using machine learning.
• Sentiment Analysis: Detects emotional tone (positive, neutral, or negative) in text.
• Trend Prediction: Predicts future trends in public opinion based on historical data.

Installation & Setup

1. Install the necessary environment dependencies (optional):

   conda install --file requirements.txt
   

2. Configure your MySQL database:

   • Run createTables.sql to set up the required tables.
   • Modify the MySQL configuration in the program accordingly.

3. Start the project with Flask:

   python app.py