extracts bigrams from a text using log likelihood
The program takes two arguments:
- inputfilename: The name of the text file to process
- outputfilename: The name of the file to write the results.
python ll.py inputfilename outputfilename
For example,
python ll.py lotusSutra.txt results.txt
The programs reads the configuration file, `ll.conf``.
The format of the configuration file is
IgnoreCase = no
NumResults = 20
Punctuation ="“”.,?:;-"'
If the value of IgnoreCase is yes then all words are lowercased. As a result The Ohio, the Ohio, and the ohio are all considered the same. If the value is no all those are considered unique.
The values can be either the word all or a number, This indicates the number of words to print to the result file.
All the listed punctuation will be stripped from both the beginnings and ends of words. For example, using the value shown above -ice-cream- will be considered as ice-cream, and end.)" will be considered end