CS6200-Project

=============================================================================== SETUP:

2. Import the project into one of the IDEs (Eclipse, Intellij) and run Index.java:
   a) Import the project into your IDE.
   b) Run Index.java as a Java application.
   

=============================================================================== Flowchart of Index.java

1. Phase 1: Indexing, retrieval systems, Pesudo Relevance Feedback, Baseline runs using stop word list and Stemmed corpus.
   1. Task 1: Indexing and retrieval systems
      1. Parse the documents and create a word unigram inverted Index
      2. Perform baseline (BM25, tf-idf, smoothed query likelihood and lucene) run
         1. BM25 retrieval model

 1. Task 1: Indexing, retrieval systems.
   1. Parse the documents and create a a word unigram inverted index
   2. Perform baseline (BM25, tf-idf, smoothed query likelihood and lucene) runs.
     1. BM25 retrieval model.
     2. TF-IDF retrieval model.
     3. Smoothed Query likelihood retrieval model.
     4. Lucene retrieval model.
           2. Task 2: Pesudo Relevance feedback (Note: Smoothed query lilihood retrieval system is used).
   3. Task 3: Baseline runs using stopp word list and Stemmed corpus.
          2. Phase 2: Snippet Generation on the results of smoothed query likelihood.
 3. Phase 3: Evaluation on 4 baseline runs, 1 query refinement run (PRF) and 3 stopped runs.