def buildTrie():
trie = Trie()
for professor in Professor.get_all_professors():
for str in formatSearchContent(professor.name):
trie.add(str, professor.get_id())
for str in formatSearchContent(professor.title):
trie.add(str, professor.get_id())
if professor.special_title:
for str in formatSearchContent(professor.special_title):
trie.add(str, professor.get_id())
if professor.introduction:
for str in formatSearchContent(professor.introduction):
trie.add(str, professor.get_id())
for research_area in professor.research_areas:
for str in formatSearchContent(research_area):
trie.add(str, professor.get_id())
for research_interest in professor.research_interests:
for str in formatSearchContent(research_interest):
trie.add(str, professor.get_id())
for research_group in professor.research_groups:
for str in formatSearchContent(research_group):
trie.add(str, professor.get_id())
if professor.office:
for str in formatSearchContent(professor.office):
trie.add(str, professor.get_id())
if professor.phone:
for str in formatSearchContent(professor.phone):
trie.add(str, professor.get_id())
if professor.email:
for str in formatSearchContent("".join(re.findall(r'(.+)@', professor.email))):
trie.add(str, professor.get_id())
return trie
def search_professors(query_words):
query_words = formatSearchContent(query_words)
pid = []
relevence = {}
trie = PROFESSOR_TRIE
for word in query_words:
pids_of_word = set()
for string in trie.searchSubstring(word):
pids_of_string = trie.get(string)
for prof_id in pids_of_string:
if not(prof_id in relevence):
relevence[prof_id] = 0.0
relevence[prof_id] = relevence[prof_id] + float(len(word))/len(string) * OCCURRENCE_DICT[prof_id][string]
pids_of_word = pids_of_word.union(pids_of_string)
pid.append(pids_of_word)
ret = [Professor.get_professor(prof_id) for prof_id in reduce(lambda x, y: x.intersection(y), pid)]
return sorted(ret, key=lambda x: -relevence[x.key.id()])
def buildOccurrenceDict():
ret = dict()
for professor in Professor.get_all_professors():
prof_dict = dict()
for str in formatSearchContent(professor.name):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
for str in formatSearchContent(professor.title):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
if professor.special_title:
for str in formatSearchContent(professor.special_title):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
if professor.introduction:
for str in formatSearchContent(professor.introduction):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
for research_area in professor.research_areas:
for str in formatSearchContent(research_area):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
for research_interest in professor.research_interests:
for str in formatSearchContent(research_interest):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
for research_group in professor.research_groups:
for str in formatSearchContent(research_group):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
if professor.office:
for str in formatSearchContent(professor.office):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
if professor.phone:
for str in formatSearchContent(professor.phone):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
if professor.email:
for str in formatSearchContent("".join(re.findall(r'(.+)@', professor.email))):
if str in prof_dict:
prof_dict[str] = prof_dict[str] + 1
else:
prof_dict[str] = 0
ret[professor.get_id()] = prof_dict
return ret
