PROJECT TITLE :

Classification and Summarization for Informative Tweets

ABSTRACT:

Microblogging services such as Twitter, Facebook, and others have evolved into a significant platform for people to express their thoughts, desires, and other needs. Users can utilize it to send short messages to their internet audience.

These messages, which can include photographs, videos, or voice notes, are a hybrid of blogging and minute messaging. For obtaining real-time informational data, we have principally concentrated on information provided by microblogging sites.

Microblogging websites are frequently utilized by people all over the world to document what is going on in their daily lives. As a result, data collected through these sites eventually aids us in obtaining non-manipulated data directly from users. A disaster dataset (Fani Cyclone dataset) is explored in this research, which comprises of tweets about a Cyclone dubbed "Fani." The tweets are pre-processed before being divided into informative and non-informative categories. When pre-processed data is taken into account, we were able to attain a classification accuracy of 74:268 percent.

Because we're dealing with a disaster dataset, we've compiled a list of useful tweets for the relevant authorities, which will aid them in getting a better understanding of the data.