The purpose of this program is twofold:

• Calculate parameters for a known, translated language pair.
• Infer the language pair for an unknown, translated language pair.

To calculate the parameters for a known, translated language pair:
python main.py --param [--batch] --source <language_corpus.txt> --target <language_corpus.txt> [--groups <int,int,int>] [--plot]

To infer the language pair for an unknown, translated language pair:
python main.py --infer [--batch] --source <language_corpus.txt> --target <language_corpus.txt> [--groups <int,int,int>] [--plot]

--source: The source language corpus file. The number of lines in this file must match the target language file.

--target: The target language corpus file. The number of lines in this file must match the source language file.

--batch: Starts n proceses, where n is the number of the computer's available CPUs on which this program is run, to speed up the time it takes to calculate the edit distances for a language pair. It may be used for either --param or --infer.

--param: Calculates the edit distances for a known, translated language pair specified by --source and --target. The edit distances are used to calaculate four parameters, explained in more detail in the README's section Parameters. The parameters are then written to a file parameters.txt.

--infer: Calculates the edit distances for an unkown, translated language pair specified by --source and --target. The edit distances are used to calculate four parameters, which are compared against the parameters for a known set of translated languages. The inferred probability for each language is written to a file inferences.txt.

--groups: Comma separated integers by which the edit distances will be grouped and counted. For example, if you specify the numbers 5,10,15,25,100, then all edit distances of five or less will be counted in group 5, all edit distances of 10 or less will be counted in group 10, and so on. NOTE: no spaces between the comma separated integers!

--plot: Plots the edit distances calculated for the source and target languages. A link to the charts on plot.ly should be automatically generated. Please ignore the UserWarnings posted by the plot.ly API.

The four parameters calculated by this program are listed below. More information about the parameters may be found in the PDF entitled 'Project Details':

• Worst Expanding Factor
• Worst Contracting Factor
• Worst Stretching Factor
• Average Stretching Factor

If you choose to generate a graph, please note that a maximum of 25 graphs may be stored at any given time. If you exceed this limit, please inform me so that I may delete one of the older graphs are your request.

• Python2.7: yes, it is supported
• Python3: unknown, there is no guarantee the program will work because it was neither tested nor run for this version.

• nltk: used to calculate edit distance
• plotly: used to plot a graph via https://plot.ly/
• linecache: used to improve performance when reading text from file