def rank(self, query):
query = StemmerHelper.stem_text(query)
term_freqs = self.__get_tf(query)
query_vector = {}
for term, tf in term_freqs.iteritems():
idf = 0.0
if self.__idfs.has_key(term):
idf = self.__idfs[term]
tfidf = tf * idf
print "{0} tf {1} : idf {2} : {3}".format(term, tf, idf, tfidf )
query_vector[term] = tfidf
union = set()
for term in term_freqs.keys():
if self.__inverted_file.has_key(term):
union = union | self.__inverted_file[term]
scores = []
for id in union:
scores.append({'id': id, 'score': self.__similarity(query_vector, self.__get_document_vector(id))})
sorted_scores = sorted(scores, key=lambda k: k['score'], reverse=True)
self.__search_result = []
for score in sorted_scores:
document = self.__documents[score['id']]
document.set_score(score['score'])
self.__search_result.append(document)
return self.__search_result
def __init_keywords(self, keywords_filename):
print keywords_filename
self.__documents = []
#p = PorterStemmer()
file = open(keywords_filename)
try:
line = file.readline()
while line:
stemmed_word = StemmerHelper.stem_text(line)
if len(stemmed_word) > 0:
self.__keywords.add(stemmed_word)
line = file.readline()
except IOError:
print "Error: can\'t find keywords file or read data"
finally:
file.close()
return
def __init__(self, title, text):
self.__text = text
self.__stemmed_text = StemmerHelper.stem_text(text)
self.__title = title
self.__score = 0.0