System for monitoring natural disasters using natural language processing in the social network Twitter

This paper presents the design and implementation of an automated system that allows monitoring the social network Twitter, making a connection to the API, to filter content according to four categories (volcanic, telluric, fires and climatological) which affect Ecuador because of its geographical location, taking into account that these cannot be easily predicted, and stores all tweets in a database for analysis. The filtering process is performed by using the NLTK tool with which the frequency of a word is determined within a tweet, to be classified later in one of the proposed categories. The results for each category are displayed on a web page that contains real-time statistics of the database. This work provides access to information on natural disasters because they are classified.