def DetectNonOutliers(listtuple_pred_true_text):
printClusterEvaluation_list(listtuple_pred_true_text)
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
print("true clusters=" +
str(len(groupTxtByClass(listtuple_pred_true_text, True))))
#ComputePurity(dic_tupple_class)
totalItems = 0
itemsInClusterList = []
for label, pred_true_txts in dic_tupple_class.items():
itemsInCluster = len(pred_true_txts)
#print("itemsInCluster="+str(itemsInCluster))
totalItems = totalItems + itemsInCluster
itemsInClusterList.append(itemsInCluster)
totalClusters = len(dic_tupple_class)
avgItemsInCluster_in_a_batch = float(totalItems) / totalClusters
std = np.std(itemsInClusterList)
print("totalItems=" + str(totalItems) + ",avgItemsInCluster_in_a_batch=" +
str(avgItemsInCluster_in_a_batch) + ",std=" + str(std))
non_outlier_pred_true_txts_in_all_clusters, outlier_pred_true_txts_in_all_clusters = DetectNonOutliersByThreshold(
dic_tupple_class, avgItemsInCluster_in_a_batch)
print("total #outliers=" +
str(len(outlier_pred_true_txts_in_all_clusters)))
#print("#non_outlier_pred_true_txts_in_all_clusters#")
#print(non_outlier_pred_true_txts_in_all_clusters)
#print("#outlier_pred_true_txts_in_all_clusters#")
#print(outlier_pred_true_txts_in_all_clusters)
#print("--Batch End--")
return [
non_outlier_pred_true_txts_in_all_clusters,
outlier_pred_true_txts_in_all_clusters, avgItemsInCluster_in_a_batch
]
def DetectNonOutliersByThreshold(dic_tupple_class, avgItemsInCluster_in_a_batch):
non_outlier_pred_true_txts_in_all_clusters=[]
outlier_pred_true_txts_in_all_clusters=[]
for label, pred_true_txts in dic_tupple_class.items():
itemsInCluster=len(pred_true_txts)
if itemsInCluster>avgItemsInCluster_in_a_batch:
#print("cluster label="+str(label)+", "+str(itemsInCluster))
textsArr=[]
for pred_true_txt in pred_true_txts:
textsArr.append(pred_true_txt[2])
vectorizer = TfidfVectorizer( max_df=1.0, min_df=1, stop_words='english', use_idf=True, smooth_idf=True, norm='l2')
x_train = vectorizer.fit_transform(textsArr)
contratio = 0.3
isf = IsolationForest(n_estimators=100, max_samples='auto', contamination=contratio, max_features=1.0, bootstrap=True, verbose=0, random_state=0, behaviour='new')
#isf=IsolationForest(n_estimators=100, max_samples='auto', contamination=contratio, max_features=1.0, bootstrap=True, verbose=0, random_state=0)
outlierPreds = isf.fit(x_train).predict(x_train)
non_outlier_pred_true_txts_in_a_cluster=[]
for i in range(len(outlierPreds)):
outlierPred=outlierPreds[i]
if outlierPred !=-1:
non_outlier_pred_true_txts_in_a_cluster.append(pred_true_txts[i])
non_outlier_pred_true_txts_in_all_clusters.append(pred_true_txts[i])
else:
outlier_pred_true_txts_in_all_clusters.append(pred_true_txts[i])
else:
non_outlier_pred_true_txts_in_all_clusters.extend(pred_true_txts)
dic_tupple_class_filteres=groupTxtByClass(non_outlier_pred_true_txts_in_all_clusters, False)
printClusterEvaluation_list(non_outlier_pred_true_txts_in_all_clusters)
print ("true clusters="+str(len(groupTxtByClass(non_outlier_pred_true_txts_in_all_clusters, True))))
#ComputePurity(dic_tupple_class_filteres)
return [non_outlier_pred_true_txts_in_all_clusters, outlier_pred_true_txts_in_all_clusters]
def GenerateTrainTest2_Percentage(percentTrainData):
trainDataRatio = 1.0
listtuple_pred_true_text = ReadPredTrueText(traintestFile)
perct_tdata = percentTrainData / 100
goodAmount_txts = int(perct_tdata *
(len(listtuple_pred_true_text) / numberOfClusters))
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
#write texts of each group in
WriteTextsOfEachGroup(textsperlabelDir, dic_tupple_class)
dic_label_outliers = Gen_WriteOutliersEachGroup(textsperlabelDir,
numberOfClusters)
train_pred_true_txts = []
test_pred_true_txts = []
for label, pred_true_txt in dic_tupple_class.items():
outlierpreds = dic_label_outliers[str(label)]
pred_true_txts = dic_tupple_class[str(label)]
if len(outlierpreds) != len(pred_true_txts):
print("Size not match for=" + str(label))
outLiers_pred_true_txt = []
count = -1
for outPred in outlierpreds:
outPred = str(outPred)
count = count + 1
if outPred == "-1":
outLiers_pred_true_txt.append(pred_true_txts[count])
test_pred_true_txts.extend(outLiers_pred_true_txt)
#remove outlierts insts from pred_true_txts
pred_true_txts_good = [
e for e in pred_true_txts if e not in outLiers_pred_true_txt
]
dic_tupple_class[str(label)] = pred_true_txts_good
for label, pred_true_txt in dic_tupple_class.items():
pred_true_txts = dic_tupple_class[str(label)]
pred_true_txt_subs = []
numTrainGoodTexts = int(perct_tdata * len(pred_true_txts))
if len(pred_true_txts) > goodAmount_txts:
pred_true_txt_subs.extend(pred_true_txts[0:goodAmount_txts])
test_pred_true_txts.extend(
pred_true_txts[goodAmount_txts:len(pred_true_txts)])
else:
pred_true_txt_subs.extend(pred_true_txts)
train_pred_true_txts.extend(pred_true_txt_subs)
trainDataRatio = len(train_pred_true_txts) / len(train_pred_true_txts +
test_pred_true_txts)
#print("trainDataRatio="+str(trainDataRatio))
if trainDataRatio <= maxTrainRatio:
WriteTrainTestInstances(trainFile, train_pred_true_txts)
WriteTrainTestInstances(testFile, test_pred_true_txts)
return trainDataRatio
def generateTrainTestTxtsByOutliers(dic_tuple_class, dic_list_outliers_class1,
maxItemsInEachClass):
trainTup_pred_true_txt = []
testTup_pred_true_txt = []
for key, value in dic_tuple_class.items():
if key not in dic_list_outliers_class1 or len(value) != len(
dic_list_outliers_class1[key]):
print("miss match=" + key)
continue
rest_traintupTPTxt = []
outliers = dic_list_outliers_class1[key]
print("collections.Counter=" + str(collections.Counter(outliers)) +
", key=" + key + ", len(outliers)=" + str(len(outliers)) +
", len(value)=" + str(len(value)))
count = -1
for tup_pred_true_text in value:
count = count + 1
#print("outliers[count]="+str(outliers[count]))
if outliers[count] == -1:
testTup_pred_true_txt.append(tup_pred_true_text)
else:
rest_traintupTPTxt.append(tup_pred_true_text)
print("len(rest_traintupTPTxt)=" + str(len(rest_traintupTPTxt)) +
",maxItemsInEachClass=" + str(maxItemsInEachClass))
if len(rest_traintupTPTxt) > maxItemsInEachClass:
trainTup_pred_true_txt.extend(
rest_traintupTPTxt[0:maxItemsInEachClass])
testTup_pred_true_txt.extend(
rest_traintupTPTxt[maxItemsInEachClass:len(rest_traintupTPTxt
)])
else:
trainTup_pred_true_txt.extend(rest_traintupTPTxt)
print("after remove outlier, max items=" + str(maxItemsInEachClass) +
", total=" +
str(len(trainTup_pred_true_txt + testTup_pred_true_txt)))
groupTxtByClass(trainTup_pred_true_txt + testTup_pred_true_txt, False)
return [trainTup_pred_true_txt, testTup_pred_true_txt]
def Evaluate(listtuple_pred_true_text):
print("evaluate total texts=" + str(len(listtuple_pred_true_text)))
preds = []
trues = []
for pred_true_text in listtuple_pred_true_text:
preds.append(pred_true_text[0])
trues.append(pred_true_text[1])
score = metrics.homogeneity_score(trues, preds)
print("homogeneity_score-whole-data: %0.8f" % score)
score = metrics.completeness_score(trues, preds)
print("completeness_score-whole-data: %0.8f" % score)
score = metrics.v_measure_score(trues, preds)
print("v_measure_score-whole-data: %0.4f" % score)
nmi_score = metrics.normalized_mutual_info_score(
trues, preds, average_method='arithmetic')
print("nmi_score-whole-data: %0.8f" % nmi_score)
# score=metrics.adjusted_mutual_info_score(trues, preds)
# print ("adjusted_mutual_info_score-whole-data: %0.4f" % score)
# score=metrics.adjusted_rand_score(trues, preds)
# print ("adjusted_rand_score-whole-data: %0.4f" % score)
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
dic_tupple_class_true = groupTxtByClass(listtuple_pred_true_text, True)
print("pred clusters=" + str(len(dic_tupple_class)) + ", true clusters=" +
str(len(dic_tupple_class_true)))
purity = ComputePurity(dic_tupple_class)
'''print("---Pred distribution")
for key,value in dic_tupple_class.items():
print(key, len(value))
print("---True distribution")
for key,value in dic_tupple_class_true.items():
print(key, len(value))'''
return [purity, nmi_score]
def EvaluateByPurity(traintestFile):
listtuple_pred_true_text = ReadPredTrueText(traintestFile)
preds = []
trues = []
for pred_true_text in listtuple_pred_true_text:
preds.append(pred_true_text[0])
trues.append(pred_true_text[1])
score = metrics.homogeneity_score(trues, preds)
print("purity_score-whole-data: %0.4f" % score)
score = metrics.normalized_mutual_info_score(trues, preds)
print("nmi_score-whole-data: %0.4f" % score)
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
ComputePurity(dic_tupple_class)
def removeOutlierConnectedComponentLexicalByItem(listtuple, batchDocs,
maxPredLabel):
outliers = []
non_outliers = []
avgItemsInCluster = 0
dic_tupple_class = groupTxtByClass(listtuple, False)
for label, items in dic_tupple_class.items():
#_components,newPred_items=clusterByConnectedComponentByItem(items)
print("to do")
return [outliers, non_outliers, avgItemsInCluster, maxPredLabel]
def printClusterEvaluation_list(listtuple_pred_true_text):
preds = []
trues = []
for pred_true_text in listtuple_pred_true_text:
preds.append(pred_true_text[0])
trues.append(pred_true_text[1])
score = metrics.homogeneity_score(trues, preds)
print("homogeneity_score-whole-data: %0.4f" % score)
score = metrics.normalized_mutual_info_score(trues,
preds,
average_method='arithmetic')
print("nmi_score-whole-data: %0.4f" % score)
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
ComputePurity(dic_tupple_class)
def EvaluateByPurity(traintestFile):
listtuple_pred_true_text = ReadPredTrueText(traintestFile)
preds = []
trues = []
for pred_true_text in listtuple_pred_true_text:
preds.append(pred_true_text[0])
trues.append(pred_true_text[1])
#score = metrics.homogeneity_score(trues, preds)
#print ("homogeneity_score-whole-data: %0.4f" % score)
score = metrics.normalized_mutual_info_score(trues,
preds,
average_method='arithmetic')
#print ("nmi_score-whole-data: %0.6f" % score)
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
acc = ComputePurity(dic_tupple_class)
print("acc", acc, "nmi", score)
def comPrehensive_GenerateTrainTestTxtsByOutliersTfIDf_varoutlier(
listtuple_pred_true_text, maxItemsInEachClass, avgItemPercluster):
trainTup_pred_true_txt = []
testTup_pred_true_txt = []
dic_tuple_class = groupTxtByClass(listtuple_pred_true_text, False)
dic_list_outliers_class = {}
for key, value in dic_tuple_class.items():
txt_datas = []
for tup_pred_true_text in value:
txt_datas.append(tup_pred_true_text[2])
outlierratio = len(value) / avgItemPercluster * 0.3
print("outlierratio=" + str(outlierratio))
if outlierratio > 0.4:
outlierratio = 0.4
contratio = outlierratio #0.1
print(len(txt_datas))
vectorizer = TfidfVectorizer(max_df=1.0,
min_df=1,
stop_words='english',
use_idf=True,
smooth_idf=True,
norm='l2')
x_train = vectorizer.fit_transform(txt_datas)
isf = IsolationForest(n_estimators=100,
max_samples='auto',
contamination=contratio,
max_features=1.0,
bootstrap=True,
verbose=0,
random_state=0)
outlierPreds = isf.fit(x_train).predict(x_train)
print(len(outlierPreds))
dic_list_outliers_class[key] = outlierPreds
trainTup_pred_true_txt, testTup_pred_true_txt = generateTrainTestTxtsByOutliers(
dic_tuple_class, dic_list_outliers_class, maxItemsInEachClass)
print("#trainTup_pred_true_txt=" + str(len(trainTup_pred_true_txt)) +
", #testTup_pred_true_txt=" + str(len(testTup_pred_true_txt)))
return [trainTup_pred_true_txt, testTup_pred_true_txt]
def EvaluateByPurity(traintestFile):
listtuple_pred_true_text = ReadPredTrueText(traintestFile)
preds = []
trues = []
for pred_true_text in listtuple_pred_true_text:
preds.append(pred_true_text[0])
trues.append(pred_true_text[1])
score = metrics.homogeneity_score(trues, preds)
last_homoginity=score
print ("homogeneity_score-whole-data: %0.4f" % score)
#score = metrics.normalized_mutual_info_score(trues, preds, average_method='arithmetic')
score = metrics.normalized_mutual_info_score(trues, preds)
last_nmi=score
print ("nmi_score-whole-data: %0.4f" % score)
last_completeness=metrics.completeness_score(trues, preds)
print("completeness="+str(last_completeness))
dic_tupple_class=groupTxtByClass(listtuple_pred_true_text, False)
return ComputePurity(dic_tupple_class)
def outlierBySmallestGroups(newPred_OldPred_true_texts):
outliersInCluster = []
non_outliersInCluster = []
rows = len(newPred_OldPred_true_texts)
#np.concatenate((A[:,0].reshape(2,1), A[:,2:4]),axis=1)
np_arr = np.array(newPred_OldPred_true_texts)
newPred_true_texts = np.concatenate(
(np_arr[:, 0].reshape(rows, 1), np_arr[:, 2:4]), axis=1).tolist()
dic_tupple_class = groupTxtByClass(newPred_true_texts, False)
maxGroupSize = -10
newPredMaxLabel = ""
totalGroups = len(dic_tupple_class)
#print("totalGroups by connComp="+str(totalGroups))
for label, pred_true_txts in dic_tupple_class.items():
groupSize = len(pred_true_txts)
if maxGroupSize < groupSize:
newPredMaxLabel = label
maxGroupSize = groupSize
for newPred_OldPred_true_text in newPred_OldPred_true_texts:
newPredLabel = newPred_OldPred_true_text[0]
oldPredLabel = newPred_OldPred_true_text[1]
trueLabel = newPred_OldPred_true_text[2]
text = newPred_OldPred_true_text[3]
OldPred_true_text = [oldPredLabel, trueLabel, text]
if str(newPredLabel) == str(newPredMaxLabel):
non_outliersInCluster.append(OldPred_true_text)
else:
outliersInCluster.append(OldPred_true_text)
'''if totalGroups>1:
#split the group into outlier/non-outliers"
non_outliersInCluster=dic_tupple_class[newPredMaxLabel]
for label, pred_true_txts in dic_tupple_class.items():
if label!=newPredMaxLabel:
outliersInCluster.extend(pred_true_txts)
else:
non_outliersInCluster=newPred_OldPred_true_texts'''
return [outliersInCluster, non_outliersInCluster]
def ComputePurity(dic_tupple_class):
totalItems = 0
maxGroupSizeSum = 0
for label, pred_true_txts in dic_tupple_class.items():
totalItems = totalItems + len(pred_true_txts)
dic_tupple_class_originalLabel = groupTxtByClass(pred_true_txts, True)
maxMemInGroupSize = minIntVal
maxMemOriginalLabel = ""
for orgLabel, org_pred_true_txts in dic_tupple_class_originalLabel.items(
):
if maxMemInGroupSize < len(org_pred_true_txts):
maxMemInGroupSize = len(org_pred_true_txts)
maxMemOriginalLabel = orgLabel
maxGroupSizeSum = maxGroupSizeSum + maxMemInGroupSize
acc = maxGroupSizeSum / totalItems
#print("acc whole data="+str(acc))
return acc
def ComputePurity(dic_tupple_class):
totalItems=0
maxGroupSizeSum =0
for label, pred_true_txts in dic_tupple_class.items():
totalItems=totalItems+len(pred_true_txts)
dic_tupple_class_originalLabel=groupTxtByClass(pred_true_txts, True)
maxMemInGroupSize=minIntVal
maxMemOriginalLabel=""
for orgLabel, org_pred_true_txts in dic_tupple_class_originalLabel.items():
if maxMemInGroupSize < len(org_pred_true_txts):
maxMemInGroupSize=len(org_pred_true_txts)
maxMemOriginalLabel=orgLabel
maxGroupSizeSum=maxGroupSizeSum+maxMemInGroupSize
purity=maxGroupSizeSum/totalItems
print("purity majority whole data="+str(purity))
last_purity_custom=purity
return purity
def extrcatLargeClusterItems(listtuple):
dic_tupple_class = groupTxtByClass(listtuple, False)
itemCounts = []
items_to_cluster = []
items_to_not_cluster = []
for label, tuples in dic_tupple_class.items():
#if len(tuples)<3:
# continue
itemCounts.append(len(tuples))
std = statistics.stdev(itemCounts)
mean = statistics.stdev(itemCounts)
for label, tuples in dic_tupple_class.items():
no_items = len(tuples)
if no_items >= mean + 1.2 * std:
items_to_cluster.extend(tuples)
else:
items_to_not_cluster.extend(tuples)
return [items_to_cluster, items_to_not_cluster]
def ComputePurity(dic_tupple_class):
totalItems = 0
maxGroupSizeSum = 0
for label, pred_true_txts in dic_tupple_class.items():
totalItems = totalItems + len(pred_true_txts)
# print("pred label="+label+", #texts="+str(len(pred_true_txts)))
dic_tupple_class_originalLabel = groupTxtByClass(pred_true_txts, True)
maxMemInGroupSize = -1000000
maxMemOriginalLabel = ""
for orgLabel, org_pred_true_txts in dic_tupple_class_originalLabel.items(
):
# print("orgLabel label="+orgLabel+", #texts="+str(len(org_pred_true_txts)))
if maxMemInGroupSize < len(org_pred_true_txts):
maxMemInGroupSize = len(org_pred_true_txts)
maxMemOriginalLabel = orgLabel
# print("\n")
# print(str(label)+" purity="+str(maxMemInGroupSize/len(pred_true_txts))+", items="+str(len(pred_true_txts))+", max match#="+str(maxMemInGroupSize))
# print_by_group(pred_true_txts)
maxGroupSizeSum = maxGroupSizeSum + maxMemInGroupSize
purity = maxGroupSizeSum / float(totalItems)
print("purity majority whole data=" + str(purity))
return purity
def EvaluateByPurity(traintestFile):
listtuple_pred_true_text = ReadPredTrueText(traintestFile)
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
ComputePurity(dic_tupple_class)
def ClusterByHDbScan(listtuple_pred_true_text, avgItemsInCluster_in_a_batch):
print("\nClusterByHDbScan")
printClusterEvaluation_list(listtuple_pred_true_text)
print(len(listtuple_pred_true_text), avgItemsInCluster_in_a_batch)
dic_tupple_class_predicted = groupTxtByClass(listtuple_pred_true_text,
False)
numberOfClusters_predicted = len(dic_tupple_class_predicted)
dic_tupple_class_true = groupTxtByClass(listtuple_pred_true_text, True)
numberOfClusters_true = len(dic_tupple_class_true)
print("numberOfClusters_true=" + str(numberOfClusters_true) +
", numberOfClusters_predicted=" + str(numberOfClusters_predicted))
train_data = []
train_predlabels = []
train_trueLabels = []
for pred_true_text in listtuple_pred_true_text:
train_predlabels.append(pred_true_text[0])
train_trueLabels.append(pred_true_text[1])
train_data.append(pred_true_text[2])
vectorizer = TfidfVectorizer(max_df=1.0,
min_df=1,
stop_words='english',
use_idf=True,
smooth_idf=True,
norm='l2')
X = vectorizer.fit_transform(train_data)
svd = TruncatedSVD(2)
normalizer = Normalizer(copy=False)
lsa = make_pipeline(svd, normalizer)
X_svd = lsa.fit_transform(X)
min_cluster_size_in_a_batch = int(math.ceil(avgItemsInCluster_in_a_batch))
min_cluster_size_in_a_batch = 2
clusterer = hdbscan.HDBSCAN(min_cluster_size=min_cluster_size_in_a_batch)
clusterer.fit(X)
X_hdbscan_labels = clusterer.labels_
print("X-total-clusters=" + str(X_hdbscan_labels.max()))
print("Homogeneity: %0.4f" %
metrics.homogeneity_score(train_trueLabels, X_hdbscan_labels))
print("Completeness: %0.4f" %
metrics.completeness_score(train_trueLabels, X_hdbscan_labels))
print("V-measure: %0.4f" %
metrics.v_measure_score(train_trueLabels, X_hdbscan_labels))
print("Adjusted Rand-Index: %.4f" %
metrics.adjusted_rand_score(train_trueLabels, X_hdbscan_labels))
print("nmi_score-whole-data: %0.4f" %
metrics.normalized_mutual_info_score(
train_trueLabels, X_hdbscan_labels, average_method='arithmetic'))
clusterer_svd = hdbscan.HDBSCAN(
min_cluster_size=min_cluster_size_in_a_batch)
clusterer_svd.fit(X_svd)
X_svd_hdbscan_labels = clusterer_svd.labels_
db = DBSCAN().fit(X_svd)
X_svd_dbscan_labels = db.labels_
print("X-svd-total-clusters=" + str(X_svd_hdbscan_labels.max()))
print("Homogeneity: %0.4f" %
metrics.homogeneity_score(train_trueLabels, X_svd_hdbscan_labels))
print("Completeness: %0.4f" %
metrics.completeness_score(train_trueLabels, X_svd_hdbscan_labels))
print("V-measure: %0.4f" %
metrics.v_measure_score(train_trueLabels, X_svd_hdbscan_labels))
print("Adjusted Rand-Index: %.4f" %
metrics.adjusted_rand_score(train_trueLabels, X_svd_hdbscan_labels))
print("nmi_score-whole-data: %0.4f" %
metrics.normalized_mutual_info_score(train_trueLabels,
X_svd_hdbscan_labels,
average_method='arithmetic'))
print("X-svd-dbscan-total-clusters=" + str(X_svd_dbscan_labels.max()))
print("Homogeneity: %0.4f" %
metrics.homogeneity_score(train_trueLabels, X_svd_dbscan_labels))
print("Completeness: %0.4f" %
metrics.completeness_score(train_trueLabels, X_svd_dbscan_labels))
print("V-measure: %0.4f" %
metrics.v_measure_score(train_trueLabels, X_svd_dbscan_labels))
print("Adjusted Rand-Index: %.4f" %
metrics.adjusted_rand_score(train_trueLabels, X_svd_dbscan_labels))
print("nmi_score-whole-data: %0.4f" %
metrics.normalized_mutual_info_score(train_trueLabels,
X_svd_dbscan_labels,
average_method='arithmetic'))
def clusteringDCT(pred_true_txt_ind_prevPreds, wordVectorsDic, batchDocs,
maxPredLabel):
print("#m-stream-cleaned")
Evaluate(pred_true_txt_ind_prevPreds)
pred_true_text_ind_prevPreds_to_cluster, pred_true_text_ind_prevPreds_to_not_cluster = extrcatLargeClusterItems(
pred_true_txt_ind_prevPreds)
print("3 rd=" + str(pred_true_text_ind_prevPreds_to_cluster[0][3]))
print("4 rd=" + str(pred_true_text_ind_prevPreds_to_cluster[0][4]))
'''minPredToC, maxPredToC, minTrueToC, maxTrueToC=findMinMaxLabel(pred_true_text_ind_prevPreds_to_cluster)
print("minPred, maxPred, minTrue, maxTrue=(pred_true_text_ind_prevPreds_to_cluster)")
print(minPredToC, maxPredToC, minTrueToC, maxTrueToC)
minPredToNC, maxPredToNC, minTrueToNC, maxTrueToNC=findMinMaxLabel(pred_true_text_ind_prevPreds_to_not_cluster)
print("minPred, maxPred, minTrue, maxTrue=(pred_true_text_ind_prevPreds_to_not_cluster)")
print(minPredToNC, maxPredToNC, minTrueToNC, maxTrueToNC)'''
all_pred_clusters = len(groupTxtByClass(pred_true_txt_ind_prevPreds,
False))
pred_clusters = len(
groupTxtByClass(pred_true_text_ind_prevPreds_to_cluster, False))
non_pred_clusters = len(
groupTxtByClass(pred_true_text_ind_prevPreds_to_not_cluster, False))
print("#clusters=" + str(pred_clusters))
print("#not clusters=" + str(non_pred_clusters))
print("this clustering with embedding DCT")
pred_clusters = non_pred_clusters - pred_clusters
print("#update clusters=" + str(pred_clusters))
nparr = np.array(pred_true_text_ind_prevPreds_to_cluster)
print("3 rd=" + str(pred_true_text_ind_prevPreds_to_cluster[0][3]))
print("4 rd=" + str(pred_true_text_ind_prevPreds_to_cluster[0][4]))
preds = list(nparr[:, 0])
trues = list(nparr[:, 1])
texts = list(nparr[:, 2])
inds = list(nparr[:, 3])
prevPreds = list(nparr[:, 4])
skStopWords = getScikitLearn_StopWords()
texts = processTextsRemoveStopWordTokenized(texts, skStopWords)
'''dicDocFreq=getDocFreq(texts)
dctCoffs=1
X=generate_sent_vecs_toktextdata_DCT(texts, wordVectorsDic, 300,dctCoffs)
#vectorizer = TfidfVectorizer(tokenizer=stem_text,max_df=0.5,min_df=1)
#vectorizer = TfidfVectorizer(max_df=0.5,min_df=2, stop_words='english')
#X = vectorizer.fit_transform(texts)'''
'''svd = TruncatedSVD(50)
#svd = PCA(n_components=50)
normalizer = Normalizer(copy=False)
lsa = make_pipeline(svd, normalizer)
#X=X.toarray()
X = lsa.fit_transform(X)'''
'''km = KMeans(n_clusters=pred_clusters, init='k-means++', max_iter=100,random_state=0)
km.fit(X)
list_km_pred_true_text=combine_pred_true_txt_from_list(km.labels_, trues, texts)
print("#k-means")
Evaluate(list_km_pred_true_text)'''
'''ward = AgglomerativeClustering(n_clusters=pred_clusters, linkage='ward').fit(X)
list_hr_pred_true_text=combine_pred_true_txt_from_list(ward.labels_, trues, texts)
print("#hr-ward-DCT")
print(min(ward.labels_), max(ward.labels_))
pred_true_text_ind_prevPreds_to_not_cluster_hr=change_pred_label(pred_true_text_ind_prevPreds_to_not_cluster, pred_clusters+1)
Evaluate(list_hr_pred_true_text)
Evaluate(list_hr_pred_true_text+pred_true_text_ind_prevPreds_to_not_cluster_hr)
'''
X = generate_sent_vecs_toktextdata(texts, wordVectorsDic, 300)
ward = AgglomerativeClustering(n_clusters=pred_clusters,
linkage='ward').fit(X)
list_hr_pred_true_text_ind_prevPred = np.column_stack(
(ward.labels_, trues, texts, inds, prevPreds)).tolist()
print("#hr-ward-AVG")
pred_true_text_ind_prevPreds_to_not_cluster_hr = change_pred_label(
pred_true_text_ind_prevPreds_to_not_cluster, pred_clusters + 1)
Evaluate(list_hr_pred_true_text_ind_prevPred)
Evaluate(list_hr_pred_true_text_ind_prevPred +
pred_true_text_ind_prevPreds_to_not_cluster_hr)
#print_by_group(list_hr_pred_true_text+pred_true_text_ind_prevPreds_to_not_cluster_hr)
print("#spectral-avg")
clustering = SpectralClustering(n_clusters=pred_clusters,
assign_labels="discretize",
random_state=0).fit(X)
list_sp_pred_true_text_ind_prevPred = np.column_stack(
(clustering.labels_, trues, texts, inds, prevPreds)).tolist()
pred_true_text_ind_prevPreds_to_not_cluster_spec = change_pred_label(
pred_true_text_ind_prevPreds_to_not_cluster, pred_clusters + 1)
Evaluate(list_sp_pred_true_text_ind_prevPred)
Evaluate(list_sp_pred_true_text_ind_prevPred +
pred_true_text_ind_prevPreds_to_not_cluster_spec)
def ObtainNumberOfClusters(isByTrueLabel, listtuple_pred_true_text) : dic_tupple_class=groupTxtByClass(listtuple_pred_true_text, isByTrueLabel) num_clusters= len(dic_tupple_class) return num_clusters
def print_by_group(listtuple_pred_true_text):
dic_tupple_class = groupTxtByClass(listtuple_pred_true_text, False)
for label, pred_true_txts in sorted(dic_tupple_class.items()):
Print_list_pred_true_text(pred_true_txts)
def clusterByTfIdfFeature(list_pred_true_text):
print("pred_mstreams")
printClusterEvaluation_list(list_pred_true_text)
dic_tupple_class = groupTxtByClass(list_pred_true_text, False)
pred_clusters = len(dic_tupple_class)
print("pred_clusters for k-means=" + str(pred_clusters))
preds, trues, texts = split_pred_true_txt_from_list(list_pred_true_text)
skStopWords = getScikitLearn_StopWords()
texts = processTextsRemoveStopWordTokenized(texts, skStopWords)
vectorizer = TfidfVectorizer(tokenizer=stem_text, max_df=0.5, min_df=2)
#vectorizer = TfidfVectorizer(max_df=0.5,min_df=2, stop_words='english')
X = vectorizer.fit_transform(texts)
svd = TruncatedSVD(100)
#svd = PCA(n_components=50)
normalizer = Normalizer(copy=False)
lsa = make_pipeline(svd, normalizer)
#X=X.toarray()
X = lsa.fit_transform(X)
km = KMeans(n_clusters=pred_clusters,
init='k-means++',
max_iter=100,
random_state=0)
km.fit(X)
list_km_pred_true_text = combine_pred_true_txt_from_list(
km.labels_, trues, texts)
print("k-means")
printClusterEvaluation_list(list_km_pred_true_text)
ward = AgglomerativeClustering(n_clusters=pred_clusters,
linkage='ward').fit(X)
list_hr_pred_true_text = combine_pred_true_txt_from_list(
ward.labels_, trues, texts)
print("hr-ward")
printClusterEvaluation_list(list_hr_pred_true_text)
clustering = SpectralClustering(n_clusters=pred_clusters,
assign_labels="discretize",
random_state=0).fit(X)
list_sp_pred_true_text = combine_pred_true_txt_from_list(
clustering.labels_, trues, texts)
print("spectral")
printClusterEvaluation_list(list_sp_pred_true_text)
brc = Birch(branching_factor=50,
n_clusters=pred_clusters,
threshold=0.5,
compute_labels=True)
brc.fit_predict(X)
list_brc_pred_true_text = combine_pred_true_txt_from_list(
brc.labels_, trues, texts)
print("brc")
printClusterEvaluation_list(list_brc_pred_true_text)
gmm = GaussianMixture(n_components=pred_clusters, covariance_type='full')
gmm_labels = gmm.fit_predict(X)
list_gmm_pred_true_text = combine_pred_true_txt_from_list(
gmm_labels, trues, texts)
print("gmm")
printClusterEvaluation_list(list_gmm_pred_true_text)
