def psuedoRFSearch():
query = sys.argv[1]
df_psuedoRF = pd.DataFrame(
columns=['Doc_no', 'rank', 'Product_title', 'Result_score'])
index = MyIndexReader.MyIndexReader()
pesudo_search = PseudoRFRetrievalModel.PseudoRFRetreivalModel(index)
extractor = TransformQuery.TransformQuery()
queries = extractor.getQuries(query)
for query in queries:
print(query.queryId, "\t", query.queryContent)
results = pesudo_search.retrieveQuery(query, 20, 100, 0.4)
rank = 1
for result in results:
df_psuedoRF = df_psuedoRF.append(
{
'Doc_no': result.getDocNo(),
'rank': rank,
'Product_title': result.getDocTitle(),
'Result_score': result.getScore()
},
ignore_index=True)
rank += 1
print(df_psuedoRF)
def indexRead(term):
index = MyIndexReader.MyIndexReader()
# retrieve the token.
df = index.DocFreq(term)
ctf = index.CollectionFreq(term)
#print(" >> the token \""+term+"\" appeared in "+ str(df) +" documents and "+ str(ctf) +" times in total")
if df>0:
posting = index.getPostingList(term)
for docId in posting:
docNo = index.getDocNo(docId)
def ReadIndex(type, token):
# Initiate the index file reader.
index = MyIndexReader.MyIndexReader(type)
# retrieve the token.
df = index.DocFreq(token)
ctf = index.CollectionFreq(token)
print(" >> the token \"" + token + "\" appeared in " + str(df) +
" documents and " + str(ctf) + " times in total")
if df > 0:
posting = index.getPostingList(token)
for docId in posting:
docNo = index.getDocNo(docId)
print(docNo + "\t" + str(docId) + "\t" + str(posting[docId]))
def qrmSearch():
query = sys.argv[1]
df_qrm = pd.DataFrame(columns = ['Doc_no','rank','Product_title', 'Result_score'])
index = MyIndexReader.MyIndexReader()
search = QueryRetreivalModel.QueryRetrievalModel(index)
extractor = TransformQuery.TransformQuery()
#extractor.getQuries(query)
queries= extractor.getQuries(query)
for query in queries:
#print(query.queryId,"\t",query.queryContent)
results = search.retrieveQuery(query, 20)
rank = 1
for result in results:
df_qrm = df_qrm.append({'Doc_no': result.getDocNo(), 'rank': rank, 'Product_title': result.getDocTitle(), 'Result_score' : result.getScore()}, ignore_index=True)
rank +=1
processed_file = Path.ResultHM1 + 'result_' + user + '.txt'
names_file = Path.FilesDictDir + 'dict_' + user + '.txt'
dict_file = Path.IndexDir + "dictionary_" + user
postings_file = Path.IndexDir + "postings_" + user
dr = open(names_file, "r", encoding='cp437')
while True:
l = dr.readline()
if not l:
break
l2 = l.split(":")
doc_to_title_dict[int(l2[0])] = l2[1]
pass
index = MyIndexReader.MyIndexReader(user)
search = QueryRetreivalModel.QueryRetrievalModel(index, user)
#preprocessing and indexing ends
extractor = ExtractQuery.ExtractQuery(Query_string)
#query execution starts
query = extractor.getQuries()
results = search.retrieveQuery(query, 4, miu, doc_to_title_dict, user)
rank = 1
final_json = []
if len(results) == 0:
print("[]")
else:
for result in results: