mi_list.sort(self.cmp_mi_list, reverse=True)
return mi_list
def top_n_terms(self, n=100):
'''
Returns top n terms with highest MI
'''
top_list = []
for item in self.terms_sorted()[0:n]:
top_list.append(item['term'])
return top_list
if __name__ == '__main__':
mx = MatrixExpress()
#mx = Matrix()
mx.add_doc(doc_id='1',
doc_terms=['apple', 'mac', 'iphone', 'mac'],
doc_class='apple',
frequency=True,
do_padding=True)
mx.add_doc(doc_id='2',
doc_terms=['windows', 'word', 'excel', 'office'],
doc_class='microsoft',
frequency=True,
do_padding=True)
mx.add_doc(doc_id='3',
doc_terms=['computer', 'mac', 'iphone', 'ipad'],
doc_class='apple',
frequency=True,
return mi_list
def top_n_terms(self, n=100):
'''
Returns top n terms with highest MI
'''
top_list = []
for item in self.terms_sorted()[0:n]:
top_list.append(item['term'])
return top_list
if __name__ == '__main__':
mx = MatrixExpress()
#mx = Matrix()
mx.add_doc(doc_id='1',
doc_terms=['apple', 'mac', 'iphone', 'mac'],
doc_class= 'apple',
frequency=True, do_padding=True)
mx.add_doc(doc_id='2',
doc_terms=['windows', 'word', 'excel', 'office'],
doc_class= 'microsoft',
frequency=True, do_padding=True)
mx.add_doc(doc_id='3',
doc_terms=['computer', 'mac', 'iphone', 'ipad'],
doc_class= 'apple',
frequency=True, do_padding=True)
mx.add_doc(doc_id='4',
doc_terms=['excel', 'computer', 'office', 'xp'],
