def _init():
# terms = db.get_terms({"accFrequence": {"$gt": 300}})
terms = db.get_terms({"$or": [{"accFrequence": {"$gt": 200}}, {"queryFrequence": {"$gt": 3}}]})
for term in terms:
pinyin = term["pinyin"]
pinyinTree.insert_pinyin(pinyin)
_load_dir_post()
def _init():
#terms = db.get_terms({"accFrequence": {"$gt": 300}})
terms = db.get_terms({"$or": [{"accFrequence": {"$gt": 200}},\
{"queryFrequence": {"$gt": 3}}]})
for term in terms:
pinyin = term['pinyin']
pinyinTree.insert_pinyin(pinyin)
_load_dir_post()
def _get_related(query):
terms = db.get_terms({"queryFrequence": {"$gt": 3}})
frequences = []
minDists = []
relatedTerms = []
if not terms:
return relatedTerms
for term in terms:
t = term['_id']
if (t.find(query) != -1 or query.find(t) != -1) and t != query:
queryFrequence = term['queryFrequence']
minDist = _min_dist(query, t)
minDists.append(minDist)
frequences.append(queryFrequence)
relatedTerm = {}
relatedTerm['term'] = t
relatedTerm['dist'] = minDist
relatedTerm['fre'] = queryFrequence
relatedTerms.append(relatedTerm)
if not relatedTerms:
return []
frequences = sorted(frequences)
minDists = sorted(minDists, reverse=True)
lenOfTerm = len(relatedTerms)
related = {}
for relatedTerm in relatedTerms:
term = relatedTerm['term']
dist = relatedTerm['dist']
fre = relatedTerm['fre']
scoreOfDist = minDists.index(dist) / float(lenOfTerm)
scoreOfFre = frequences.index(fre) / float(lenOfTerm)
score = scoreOfDist * 0.4 + scoreOfFre * 0.6
related[term] = score
results = sorted(related.items(),
key=lambda related: related[1],
reverse=True)
return results[0:min(len(results), 10)]
def _get_related(query):
terms = db.get_terms({"queryFrequence": {"$gt": 3}})
frequences = []
minDists = []
relatedTerms = []
if not terms:
return relatedTerms
for term in terms:
t = term["_id"]
if (t.find(query) != -1 or query.find(t) != -1) and t != query:
queryFrequence = term["queryFrequence"]
minDist = _min_dist(query, t)
minDists.append(minDist)
frequences.append(queryFrequence)
relatedTerm = {}
relatedTerm["term"] = t
relatedTerm["dist"] = minDist
relatedTerm["fre"] = queryFrequence
relatedTerms.append(relatedTerm)
if not relatedTerms:
return []
frequences = sorted(frequences)
minDists = sorted(minDists, reverse=True)
lenOfTerm = len(relatedTerms)
related = {}
for relatedTerm in relatedTerms:
term = relatedTerm["term"]
dist = relatedTerm["dist"]
fre = relatedTerm["fre"]
scoreOfDist = minDists.index(dist) / float(lenOfTerm)
scoreOfFre = frequences.index(fre) / float(lenOfTerm)
score = scoreOfDist * 0.4 + scoreOfFre * 0.6
related[term] = score
results = sorted(related.items(), key=lambda related: related[1], reverse=True)
return results[0 : min(len(results), 10)]
def get_matches(query):
if not isinstance(query, unicode):
query = query.decode("utf-8")
string = p.get_pinyin(query, "")
matches = pinyinTree.get_match(string)
if not matches:
return None
results = {}
for match in matches:
# terms = db.get_terms({"pinyin": match, "accFrequence": {"$gt": 5}})
terms = db.get_terms({"pinyin": match})
for term in terms:
if _is_match(query, term["_id"], string, term["pinyin"]):
score = term["accFrequence"] * 0.05 + term["queryFrequence"] * 0.95
results[term["_id"]] = score
results = sorted(results.items(), key=lambda results: results[1], reverse=True)
return results[0 : min(20, len(results))]
def get_matches(query):
if not isinstance(query, unicode):
query = query.decode('utf-8')
string = p.get_pinyin(query, "")
matches = pinyinTree.get_match(string)
if not matches:
return None
results = {}
for match in matches:
#terms = db.get_terms({"pinyin": match, "accFrequence": {"$gt": 5}})
terms = db.get_terms({"pinyin": match})
for term in terms:
if _is_match(query, term['_id'], string, term['pinyin']):
score = term['accFrequence'] * 0.05 + term[
'queryFrequence'] * 0.95
results[term['_id']] = score
results = sorted(results.items(),
key=lambda results: results[1],
reverse=True)
return results[0:min(20, len(results))]
