def similarityWords(dictOxfordNouns):
print "\n --------------------------------------------------"
print "new round"
total_precision = 0;
total_recall = 0;
total_accuracy = 0;
missingWORD = 0;
total_word = 0
count_step = 0
for word in dictOxfordNouns:
# print dictOxfordNouns[word]
# if word == 'baby':
matrix_result = similarity_by_synsets_synsets_nbest_withword_average(word,dictOxfordNouns[word]);
if matrix_result == None:
missingWORD += 1;
continue
# print "\n"
print word
(precision, recall, accuracy) = CompareWithGold.compareGoldWithResult(matrix_result,word)
# print (precision, recall, accuracy)
if precision != -1:
total_precision += precision
total_recall += recall
total_accuracy += accuracy
total_word += 1
count_step += 1
# print "result -------"
# print precision
# print recall
# print accuracy
# print "\n"
else:
missingWORD += 1
# total_word = len(dictOxfordNouns) - missingWORD
precision = total_precision/total_word
recall = total_recall/total_word
accuracy = total_accuracy/total_word
print "total:"
print total_word
print precision
print recall
print accuracy
WriteParametersAndResult.append_result_to_file(precision,recall,accuracy)
return (precision, recall, accuracy)
def sim_ox_wn_via_definition_cal_word():
total_precision = 0;
total_recall = 0;
total_accuracy = 0;
total_word = 0
dict_ox = OxfordParser.get_dict_nouns()
for word in dict_ox:
if word not in __m2d_sim__:
m2d_sim = sim_ox_wn_definition(word)
__m2d_sim__[word] = m2d_sim
m2d_sim = copy.deepcopy(__m2d_sim__[word])
if m2d_sim == None or len(m2d_sim) == 0 or len(m2d_sim[0]) == 0:
continue
print word
#
# if len(m2d_sim) == 1 and len(m2d_sim[0]) == 1:
# continue
#
m2d_sim = choose_pair_0_1(m2d_sim, len(m2d_sim), len(m2d_sim[0]))
(precision, recall, accuracy) = CompareWithGold.compareGoldWithResult(m2d_sim,word)
if precision != -1:
total_precision += precision
total_recall += recall
total_accuracy += accuracy
total_word += 1
precision = total_precision/total_word
recall = total_recall/total_word
f_score = 0
if precision != 0 or recall != 0:
f_score = 2*(precision*recall)/(precision + recall)
accuracy = total_accuracy/total_word
print "total:"
print total_word
print precision
print recall
print f_score
print accuracy
Parameters.append_result_to_file( precision, recall, f_score, accuracy)
def sim_ox_wn_via_svm():
total_tp = 0.00001
total_tn = 0.00001
total_fn = 0.00001
total_fp = 0.00001
total_pair = 0
dict_ox = OxfordParser.get_dict_nouns()
flag_can_go = False
for word in dict_ox:
# if word == "brook":
# flag_can_go = True
#
# if flag_can_go == False:
# continue
word_syns_ox = dict_ox[word]
wn_synsets = WordnetHandler.get_synsets_for_word(word, "n")
m2d_sim = [[0 for x in range(len(word_syns_ox))] for x in range(len(wn_synsets))]
if len(word_syns_ox) == 1 and len(wn_synsets) == 1:
m2d_sim[0][0] = 1
else:
m2d_sim = get_m2d_sim_for_word_from_svm_result(word)
if m2d_sim == None:
continue
# DebugHandler.print_2d_matrix(m2d_sim)
m2d_sim = choose_pair_0_1(m2d_sim, len(m2d_sim), len(m2d_sim[0]))
# DebugHandler.print_2d_matrix(m2d_sim)
pair = count_pair(m2d_sim)
total_pair += pair
(tp, tn, fn, fp) = CompareWithGold.compareGoldWithResult_without_cal_result(m2d_sim,word)
if tp != -1:
total_tp += tp
total_tn += tn
total_fn += fn
total_fp += fp
precision = total_tp / (total_tp + total_fp)
recall = total_tp / (total_tp + total_fn)
accuracy = (total_tp + total_tn) / (total_tp + total_tn + total_fp + total_fn)
f_score = 0
if precision != 0 or recall != 0:
f_score = 2*(precision*recall)/(precision + recall)
print "total:"
print total_pair
print total_tp
print total_tn
print total_fn
print total_fp
print precision
print recall
print f_score
print accuracy
Parameters.append_result_to_file( precision, recall, f_score, accuracy)
current_params = Parameters.get_current_params()
current_params = copy.deepcopy(current_params)
return f_score, current_params
def sim_ox_wn_via_definition_morpho_cal_syns():
total_tp = 0.00001
total_tn = 0.00001;
total_fn = 0.00001;
total_fp = 0.00001;
total_pair = 0
dict_ox = OxfordParser.get_dict_nouns()
for word in dict_ox:
# if word != 'blaze':
# continue
#
if word not in __m2d_sim__:
m2d_sim = sim_ox_wn_definition(word)
__m2d_sim__[word] = m2d_sim
m2d_sim = copy.deepcopy(__m2d_sim__[word])
if m2d_sim == None or len(m2d_sim) == 0 or len(m2d_sim[0]) == 0:
continue
# if len(m2d_sim) == 1 or len(m2d_sim[0]) == 1:
# continue
if word not in __dict_ngrams__:
m2d_jacc = __m2d_sim_jacc__[word]
m2d_2grams = __m2d_sim_2grams__[word]
m2d_3grams = __m2d_sim_3grams__[word]
m2d_4grams = __m2d_sim_4grams__[word]
# DebugHandler.print_2d_matrix(m2d_jacc)
# DebugHandler.print_2d_matrix(m2d_2grams)
# DebugHandler.print_2d_matrix(m2d_3grams)
# DebugHandler.print_2d_matrix(m2d_4grams)
#
m2d_ngrams = [[0 for x in range(len(m2d_sim[0]))] for x in range(len(m2d_sim))]
monogram_weight = 0.25
bigram_weight = 0.25
trigram_weight = 0.25
for i in range(len(m2d_sim)):
for j in range(len(m2d_sim[0])):
m2d_ngrams[i][j] = m2d_jacc[i][j]*monogram_weight \
+ m2d_2grams[i][j]*bigram_weight \
+ m2d_3grams[i][j]*(trigram_weight) \
+ m2d_4grams[i][j]*(1- monogram_weight - bigram_weight - trigram_weight)
__dict_ngrams__[word] = m2d_ngrams
# print word
m2d_ngrams = __dict_ngrams__[word]
# DebugHandler.print_2d_matrix(m2d_ngrams)
# DebugHandler.print_2d_matrix(m2d_sim)
# ngram_weight = 0.075
# for iWnWord in range(len(m2d_sim)):
# for iDictWord in range(len(m2d_sim[0])):
# jacc = m2d_jacc[iWnWord][iDictWord]
# ngrams = m2d_ngrams[iWnWord][iDictWord]
# m2d_jacc[iWnWord][iDictWord] = jacc*(1-ngram_weight) + ngrams*ngram_weight
#
JACCARD_WEIGHT = Parameters.MORPHO.JACCARD
for i in range(len(m2d_sim)):
for j in range(len(m2d_sim[0])):
m2d_sim[i][j] = m2d_sim[i][j]*(1-JACCARD_WEIGHT) + JACCARD_WEIGHT*(m2d_ngrams[i][j]);
# DebugHandler.print_2d_matrix(m2d_sim)
# if len(m2d_sim) == 1 and len(m2d_sim[0]) == 1:
# continue
#
m2d_sim = choose_pair_0_1(m2d_sim, len(m2d_sim), len(m2d_sim[0]))
# m2d_sim = pair_0_1_reducing_m2d_sim(m2d_sim, len(m2d_sim), len(m2d_sim[0]), word)
# print word
# DebugHandler.print_2d_matrix(m2d_sim)
pair = count_pair(m2d_sim)
total_pair += pair
(tp, tn, fn, fp) = CompareWithGold.compareGoldWithResult_without_cal_result(m2d_sim,word)
# precision = tp / (tp + fp + 0.0001)
# recall = tp / (tp + fn + 0.0001)
# accuracy = (tp + tn) / (tp + tn + fp + fn + 0.0001)
#
# f_score = 0
# if precision != 0 or recall != 0:
# f_score = 2*(precision*recall)/(precision + recall)
# if f_score < 0.5:
# print word
# print f_score
# print tp
# print tn
# print fn
# print fp
#
if tp != -1:
total_tp += tp
total_tn += tn
total_fn += fn
total_fp += fp
precision = total_tp / (total_tp + total_fp)
recall = total_tp / (total_tp + total_fn)
accuracy = (total_tp + total_tn) / (total_tp + total_tn + total_fp + total_fn)
f_score = 0
if precision != 0 or recall != 0:
f_score = 2*(precision*recall)/(precision + recall)
print "total:"
print total_pair
print total_tp
print total_tn
print total_fn
print total_fp
print precision
print recall
print f_score
print accuracy
Parameters.append_result_to_file( precision, recall, f_score, accuracy)
current_params = Parameters.get_current_params()
current_params = copy.deepcopy(current_params)
return f_score, current_params
def sim_ox_wn_via_definition_cal_syns():
total_tp = 0.;
total_tn = 0.;
total_fn = 0.0;
total_fp = 0.0;
total_pair = 0
dict_ox = OxfordParser.get_dict_nouns()
for word in dict_ox:
# if word != 'bank':
# continue
#
if word not in __m2d_sim__:
m2d_sim = sim_ox_wn_definition(word)
__m2d_sim__[word] = m2d_sim
m2d_sim = copy.deepcopy(__m2d_sim__[word])
if m2d_sim == None or len(m2d_sim) == 0 or len(m2d_sim[0]) == 0:
continue
# if len(m2d_sim) == 1 and len(m2d_sim[0]) == 1:
# continue
#
m2d_sim = choose_pair_0_1(m2d_sim, len(m2d_sim), len(m2d_sim[0]))
# m2d_sim = pair_0_1_reducing_m2d_sim(m2d_sim, len(m2d_sim), len(m2d_sim[0]), word)
# print word
pair = count_pair(m2d_sim)
total_pair += pair
(tp, tn, fn, fp) = CompareWithGold.compareGoldWithResult_without_cal_result(m2d_sim,word)
if tp != -1:
total_tp += tp
total_tn += tn
total_fn += fn
total_fp += fp
precision = total_tp / (total_tp + total_fp)
recall = total_tp / (total_tp + total_fn)
accuracy = (total_tp + total_tn) / (total_tp + total_tn + total_fp + total_fn)
f_score = 0
if precision != 0 or recall != 0:
f_score = 2*(precision*recall)/(precision + recall)
print "total:"
print total_pair
print total_tp
print total_tn
print total_fn
print total_fp
print precision
print recall
print f_score
print accuracy
Parameters.append_result_to_file( precision, recall, f_score, accuracy)
current_params = Parameters.get_current_params()
current_params = copy.deepcopy(current_params)
return f_score, current_params
