def __init__(self, vec_file, pap, pat, pro):
# self.wm = gensim.models.KeyedVectors.load_word2vec_format(vec_file,binary=True)
self.wm = gensim.models.word2vec.Word2Vec.load_word2vec_format(
vec_file, binary=True)
self.paper_index = AnnoyIndexer()
self.paper_index.load(pap)
self.patent_index = AnnoyIndexer()
self.patent_index.load(pat)
self.project_index = AnnoyIndexer()
self.project_index.load(pro)
self.t2v = Convert2Vec(self.wm)
self.cuttor = FilterCut()
self.db = DB()
self.featureIndex = self.buildFeatureIndex()
def get_author_by_sql(self, typee, ids):
db = DB()
return db.getAuthors(typee, ids) # 使用MySQL获取信息
def expertDocsSort(self, expertId, txt, topN):
vec = self.t2v.text2v(txt, self.cuttor)
annoy = AnnoyIndex(200)
count = 0
annoy.add_item(count, vec)
count = count + 1
db = DB()
papers = db.getPapers(expertId)
for p in papers:
p[3] = self.t2v.text2v(p[1] + p[2], self.cuttor)
if p[3] is not None:
annoy.add_item(count, p[3])
p[3] = annoy.get_distance(0, count)
count = count + 1
papers = sorted(papers, key=lambda p: p[3])
papersFormated = []
for p in papers:
if len(papersFormated) == topN:
break
map = {}
if p[0] is not None:
map['paperId'] = p[0].encode('utf8')
else:
map['paperId'] = p[0]
if p[1] is not None:
map['name'] = p[1].encode('utf8')
else:
map['name'] = p[1]
if p[4] is not None:
map['authors'] = p[4].encode('utf8')
else:
map['authors'] = p[4]
if p[5] is not None:
map['journalName'] = p[5].encode('utf8')
else:
map['journalName'] = p[5]
if p[6] is not None:
map['year'] = p[6].encode('utf8')
else:
map['year'] = p[6]
papersFormated.append(map)
count = 0
annoy.unload()
annoy.add_item(count, vec)
count = count + 1
patents = db.getPatents(expertId)
for p in patents:
p[3] = self.t2v.text2v(p[1] + p[2], self.cuttor)
if p[3] is not None:
annoy.add_item(count, p[3])
p[3] = annoy.get_distance(0, count)
count = count + 1
patents = sorted(patents, key=lambda p: p[3])
patentsFormated = []
for p in patents:
if len(patentsFormated) == topN:
break
map = {}
if p[0] is not None:
map['patentId'] = p[0].encode('utf8')
else:
map['patentId'] = p[0]
if p[4] is not None:
map['publicationNo'] = p[4].encode('utf8')
else:
map['publicationNo'] = p[4]
if p[1] is not None:
map['name'] = p[1].encode('utf8')
else:
map['name'] = p[1]
if p[5] is not None:
map['inventors'] = p[5].encode('utf8')
else:
map['inventors'] = p[5]
if p[6] is not None:
map['applicant'] = p[6].encode('utf8')
else:
map['applicant'] = p[6]
if p[7] is not None:
map['year'] = p[7].encode('utf8')
else:
map['year'] = p[7]
patentsFormated.append(map)
count = 0
annoy.unload()
annoy.add_item(count, vec)
count = count + 1
projects = db.getProjects(expertId)
for p in projects:
p[3] = self.t2v.text2v(p[1] + p[2], self.cuttor)
if p[3] is not None:
annoy.add_item(count, p[3])
p[3] = annoy.get_distance(0, count)
count = count + 1
projects = sorted(projects, key=lambda p: p[3])
projectsFormated = []
for p in projects:
if len(projectsFormated) == topN:
break
map = {}
if p[0] is not None:
map['projectId'] = p[0].encode('utf8')
else:
map['projectId'] = p[0]
if p[1] is not None:
map['name'] = p[1].encode('utf8')
else:
map['name'] = p[1]
if p[4] is not None:
map['member'] = p[4].encode('utf8')
else:
map['member'] = p[4]
if p[5] is not None:
map['unit'] = p[5].encode('utf8')
else:
map['unit'] = p[5]
if p[6] is not None:
map['year'] = p[6].encode('utf8')
else:
map['year'] = p[6]
if p[7] is not None:
map['type'] = p[7].encode('utf8')
else:
map['type'] = p[7]
projectsFormated.append(map)
result = {}
result['papers'] = papersFormated
result['patents'] = patentsFormated
result['projects'] = projectsFormated
return result
