Methodology

The hindi corpus used is the corpus used in the previous assignment. Only words with frequency of occurence less than 2 is removed. The code is run on the test file provided for hindi, and the results are compared against the GroundTruth file provided. The code aims to maximise the probability of segmentation based on the probability of unigrams.

The results are as follows:

Precision	Recall	f-score
0.9875389	0.9462686	0.9664634

Link to Code

Codes : Please supply a hindi corpus named 'hindicorpus.txt' with the format "freq word". Run the 'make' file to obtain the results.