How to develop disaster-centric search engine with Twitter data?

TwiSter is disaster-centric search engine, which source data came from Twitter live stream. The main objective is to analyse Twitter for information on disasters such as earthquake or tsunami. The three main tasks for the system is crawling, indexing and classifying.

Year 2016
Technology Machine Learning, Natural Language Processing (NLP), Web Crawling, Apache Solr, WEKA, Python, PHP, Bootstrap
Outcome TwiSter search engine can crawl up to 400K unique words and retrieve results in average of 0.015 secs