An Enhanced Framework for Disaster-Related Tweet Classification using Machine Learning Techniques

Facilities for social media have brought about a revolutionary change in the method in which individuals disseminate information. When there is a pressing need for the general people to get in touch with emergency services during a time-sensitive situation, social media has emerged as a viable alternative communication channel. Unfortunately, these agencies often find themselves inundated with information from social media, which, when combined with their limited human resources., makes it difficult for them to quickly detect, categories, and organize crucial calls for assistance. With the assistance of a sophisticated automated machine learning algorithm, disaster response groups are able to make use of this paradigm in order to get enhanced awareness of catastrophic catastrophes. The objective of this investigation is to construct a classifier based on supervised learning that is capable of reliably determining if a quantity of info (tweet) is pertaining to a genuine catastrophe or not. This will be accomplished by using the knowledge and skills that have been acquired in the fields of data investigation.