Rage Against The Lucene - IR_proj2_group14

git repository for CS7IS3 - Information Retrieval and Web Search in TCD. This project envolves a team effort in designing and implementing a Java project using the Apache Lucene library to search a large corpus of documents provided by the lecturer.

The data set contains files from:

1. The Financial Times Limited (1991, 1992, 1993, 1994)
2. The Federal Register (1994)
3. The Foreign Broadcast Information Service (1996)
4. The Los Angeles Times (1989, 1990).

Project Setup

1. Install Java 1.8, Git, Maven, lucene and trec_eval
2. Clone project
3. Compile project with: mvn clean install source:jar
4. Edit the run.sh script to select similarity and analyser models.
5. Execute the run.sh to generate results for the search engine - Results are stored in: DataSet/queryResults
6. Execute run_trec_eval.sh to generate MAP, Recall and other such metrics for the search engine. Awating qrels file for project evaluation.