gistfile1.py

import textmining
import re

f = open('patentstitles.csv','r')

def cleantext(x):
    y = []
    for line in x:
        y.append(line.split(',',1)[0])
    z = []
    for i in y:
        z.append(re.sub('/^[A-Za-z ]+$/','',i).lower())

def termdocumentmatrix(x):
    tdm = textmining.TermDocumentMatrix()
    for i in x:
        tdm.add_doc(x)
    tdm.write_csv('patenttitlematrix.csv',cutoff=2)

c = cleantext(f)
termdocumentmatrix(c)