Disaster Tweet Classification using ML
Jaydev Rathod, Girish Rathod, Pooja Upadhyay, Puja Vakhare · 2022 International Conference on Applied Artificial Intelligence and Computing (ICAAIC) · 2022
Social media platforms such as Twitter bring about a lot of details amid disaster. Today data is not only seen in the form of texts but also in the form of special characters, sound, symbols, etc on Twitter. Time is invaluable when it comes to sharing emergency information and giving the response accordingly, seen by studies. It was seen by research that twitter gets utilized effectively when it comes to disseminate useful information regarding various types of disasters like volcano eruption, flood, devastation due to geopolitical war so that concerning authority(like NDMA) could use such details to response the disasters in a very quick time and effectively. The main of the research work is achieved by using the supervised machine learning and natural language processing method system could be able to classify the tweets as disaster related tweet or non disaster related tweet. Proposed approach is based upon the use of Word2Vec & the Machine Learning Algorithm, in which feature vector is provided by Word2Vec and to classify the tweets Classification Algorithms is used.