README FILE

Assignment 1: Morphological Structure Discovery of Hindi Language
Submitted by: Priya (10535)

There are two .py files in the folder. 
1) prune.py has code for removind the words which are less thha 3 unicode characters in length from the corpus. Other pruning acctivities like removing alphanumeric words, foreign language words, less frequent words like words with fre. 1, 2 and 3 and other tasks were done using sed, awk and tr in unix.

2)code.py is the main file that needs two files to run: the hand annotated list of ground truth data named as list5.txt and the Undivided++ segmentation results named as finalSegmentation.txt The files should be kept in the same folder as the code otherwise modify the code to take file names as arguments or specify the directory path in the code.
	The functionality in the code.py is that it calculates the I,D,H for the file and finds the exact_accuracy, prercision, accuracy and F-score. There can be other uses like to see the segmentation of ground truth list by the code, we can simply write both the results(Undivided++ and our hand annotated) to the the file.
	
	In order to run the code, simply give the command: python code.py in the unix shell. The output will display on the screen.
	
For any other queries, feel free to contact me at: prigoyal@iitk.ac.in