This project aims to explore the potential of Twitter data for disaster management and risk reduction. The project includes several Python scripts that can be used for collecting, preprocessing, and analyzing Twitter data related to natural hazards and disasters.
The following .py files are included in this project:
TweetMining.py: This script can be used for collecting Twitter data using the Twitter API. It includes functions for searching for tweets based on keywords, hashtags, or user mentions, and for collecting tweets from specific locations.LocationMapping.py: This script can be used for mapping Twitter users to their geographic locations based on their profile - information. It includes functions for geocoding user locations and for visualizing the distribution of users on a map.Preprocessing.py: This script can be used for preprocessing Twitter data before conducting further analysis. It includes functions for cleaning and filtering tweets, removing URLs and special characters, and extracting relevant information such as hashtags, mentions, and URLs.HumanitarianTopics_BERT_Torch.py: This script trains and evaluates a BERT classifier to classify disaster related Tweets into Humanitarian classes.