def getNER():
if not request.get_json(force=True) or not 'text' in request.get_json(force=True):
print 'aborting'
abort(400)
# istr = request.json['text']
istr = request.get_json(force=True)['text']
n = int(request.get_json(force=True)['max_n'] )
# remove all nonAsciiCharacters
istr = removeNonAscii(istr)
inclusion_list = DB.getInclusionList()
exclusion_list = DB.getExclusionList()
istr = removeNonAscii(istr)
NER_results = NER.findNamedEntities(istr, inclusion_list)
keywords = NER_results[0:n]
# omit repeats and return lowercase
keywords = sorted(list(set(i.lower() for i in keywords)))
keywords = NER.excludeKeywords(exclusion_list, keywords)
return jsonify({'keywords': keywords}), 201
def getTFIDF():
if not request.get_json(force=True) or not 'text' in request.get_json(force=True):
print 'aborting'
abort(400)
# istr = request.json['text']
istr = request.get_json(force=True)['text']
n = int(request.get_json(force=True)['max_n'] )
# remove all nonAsciiCharacters
istr = removeNonAscii(istr)
inclusion_list = DB.getInclusionList()
exclusion_list = DB.getExclusionList()
istr = removeNonAscii(istr)
TFIDF_results = TFIDF.findTFIDFkeywords(istr)
keywords = TFIDF_results[0:n]
keywords = NER.excludeKeywords(exclusion_list, keywords)
return jsonify({'keywords': keywords}), 201
def getKeywords():
if not request.get_json(force=True) or not 'text' in request.get_json(force=True):
print 'aborting'
abort(400)
# istr = request.json['text']
istr = request.get_json(force=True)['text']
n = int(request.get_json(force=True)['max_n'] )
# remove all nonAsciiCharacters
istr = removeNonAscii(istr)
inclusion_list = DB.getInclusionList()
exclusion_list = DB.getExclusionList()
istr = removeNonAscii(istr)
NER_results = NER.findNamedEntities(istr, inclusion_list)
NER_results = list(set(NER_results))
TFIDF_results = TFIDF.findTFIDFkeywords(istr)
keywords = NER_results[0:n/2] + TFIDF_results[0:n/2]
# omit repeats and return lowercase
keywords = sorted(list(set(i.lower() for i in keywords)))
current_j = n/2
current_k = n/2
while len(keywords) < n:
if current_k < len(NER_results):
current_k += 1
if NER_results[current_k] not in keywords:
keywords.append(NER_results[current_k])
else:
current_j += 1
if TFIDF_results[current_j] not in keywords:
keywords.append(TFIDF_results[current_j])
keywords = NER.excludeKeywords(exclusion_list, keywords)
return jsonify({'keywords': keywords}), 201
import NER.extractNER as NER import TFIDF.extractTFIDF as TFIDF import DB import sample def removeNonAscii(s): return "".join(filter(lambda x: ord(x)<128, s)) inclusion_list = DB.getInclusionList() exclusion_list = DB.getExclusionList() n = 10 istr = sample.sample_text istr = removeNonAscii(istr) NER_results = NER.findNamedEntities(istr, inclusion_list) TFIDF_results = TFIDF.findTFIDFkeywords(istr) keywords = NER_results[0:n/2] + TFIDF_results[0:n/2] # omit repeats and return lowercase keywords = sorted(list(set(i.lower() for i in keywords))) keywords = NER.excludeKeywords(exclusion_list, keywords) print keywords
