Recognition of Traffic Information with the Help of Social Media Tweets

Abstract

Social media has seen a significant change in the current day in terms of how people use it to interact with friends and family, publish
content, and share it. A few social media platforms, like Twitter, LinkedIn, and Facebook, were also utilized to share the current everyday trends in the actual world. People often struggle to get to their destinations on time in major cities with heavy traffic, like Washington DC, Beijing, and New Delhi. To ensure that travelers reach their destination or locations quickly, we concentrated on collecting useful traffic-related data from social media content in our study. Additionally, we were committed to providing immediate safety precautions for traffic incidents. The key benefit is the improvement of traffic event accuracy and the presentation of location data for heavy traffic. In this study, the CNN-LSTM (Convolution Neural Network-Long Short Term Memory) approach and the Logistic Regression method were tested for their ability to extract real-time traffic information.