bawejakunal / generalised-vector-space-model

Demonstration of generalised vector space model for ranking of documents for information retrieval

Generalised Vector Space Model

Information Retrieval Project for Demonstration of Generalised Vector Space Model

Note: This project was implemented as a final assignment for CSF469 Information Retrieval at BITS Pilani K.K. Birla Goa Campus.

**GVSM Paper:**http://www.aclweb.org/anthology/E09-3009

Required environment

1. Python v3.0
2. Scikit Learn Machine Learning Library http://scikit-learn.org/

Contributors:
Kunal Baweja 2011A7PS029G
Saransh Varshneya 2011A7PS007G
Nasrin Jaleel 2011A7PS444G
Ashu Kalra 2011A7PS150G

Guidelines/Instructions to run the demo code:

1. test_large folder contains 3000 documents and test_small folder contains the small 5 files data set
2. Execute python IR_small.py for checking output on a small dataset of 5-10 documents. Please enter the path to files to be ranked in the format /path/to/current_folder/test_small/* and the path to query document as /path/to/current_folder/query.txt
3. Execute python IR_large.py for checking output on large dataset of documents. Please enter the path to files to be ranked in the format /path/to/current_folder/test_large/*. Here by default the first document read by the code is treated as the query document.

Note: The code may be tested on other datasets too provided the paths mentioned above are provided correctly by the user when prompted by the script.