This is a toy project, it indexes Gutenberg project using the standard tf-idf, and query using cosine-distance. The indexing files are less than 500 MBs compressed, 3GB uncompressed. This is a toy project of me for learning tf-idf, not production code in any level.

Run 'query_main.py' to query:

I cam across the Gutenberg-tar project where all the files are in txt format, making it very easy to process. The size is adequately big for an information retrieval project.

If you want to try indexing yourself, download the text file from http://www.gutenberg-tar.com/. Extract them at any folder and assign 'target_path' (in indexer.py) accordingly. If you want to directly query using the indexed files, you can download my index at my Google drive: https://drive.google.com/folderview?id=0B1rwTGPvWqyIbXNEenV4WDNLS00&usp=drive_web