Cross-Lingual-IR

Information retrieval system based on the Okapi BM25 model, allowing the users to search for tweets (docs.) across multiple languages.

This is the fourth (Team of four) and the final project of the Information Retrieval course (CSE 535).

1. First project dealt with collecting the tweets using the Twitter Streaming and Search API , indexing them to solr and implementing the solr features of stop word removal, language detection.
2. Second project is on implementing TAAT and DAAT techniques for the provided lucene index.
3. Third project (Team of two) is about evaluation of DFR, Okapi BM25, and Vector Space models, evaluate the models using the TREC_EVAL tool and improve the performance in terms of Mean Average Precision (MAP).

Application URL: http://chirpybird1.us-east-1.elasticbeanstalk.com