def test():
import sys
import csv
import nlp.nlp_paths as nlp_paths
f = nlp_paths.get_proj_root() + "/testing/test_interest_text"
extractor = InterestExtractor()
reader = csv.reader(open(f, "rb"), delimiter="$")
i = 0
desired = int(sys.argv[1])
for row in reader:
if len(row) == 1:
i += 1
if i == desired:
# print row[0]
print extractor.extract(row[0])
#!/usr/bin/python
#By Steve Hanov, 2011. Released to the public domain
import time
import sys
import cPickle
import nlp.nlp_paths as paths
from nltk.stem.porter import PorterStemmer
from nltk.stem.lancaster import LancasterStemmer
from nlp.util import chunk_similarity
from nltk.metrics import distance
MOST_COMMON = paths.get_proj_root()+"/util/1000_most_common.pkl";
DICTIONARY = paths.get_proj_root()+"/freebase_util/interest_lst.pkl";
CUSTOM = ['drinking', 'partying', 'wine']
TOP_K_FILTER = 200
# The Trie data structure keeps a set of words, organized with one node for
# each letter. Each node has a branch for each letter that may follow it in the
# set of words.
class TrieNode:
def __init__(self):
self.word = None
self.children = {}
def insert( self, word ):
node = self
for letter in word:
if letter not in node.children:
node.children[letter] = TrieNode()
node = node.children[letter]