1) Step1_Corpus.py was used to clean up the corpus and extract hindi character words, check their frequency and output the result in descending order of frequency in a format accepted by undivided++ 
2) start_manual.py does a very simple task of converting each line in given 300 word list in following way
		﻿4 लोकहितकारी   to  ﻿4 लोकहितकारी = ﻿लोकहितकारी
This was done so that everything during segmentation need not be typed from scratch

3) search_list30_words.py takes input as given list and cleaned corpus and creates a new file which contains proposed segmentation of algorithm found in output file of undivided++
