wordList.append(tag)
keywords2 = []
i = 0
for user in keywords:
keywords2.append([])
for word in wordList:
if word in user:
keywords2[i].append((word,1))
else:
keywords2[i].append((word,0))
i+=1
g = kmeans.open_ubigraph_server()
result,clusters = kmeans.kcluster(g,palestrantes,keywords2,k=8)
dataClusters = []
i = 0
for cluster in result:
apCount = {}
for indice in cluster:
dados = keywords2[indice]
for word,count in dados:
apCount.setdefault(word,0)
apCount[word]+= count
words = apCount.items()
words.sort(key=operator.itemgetter(1))
words.reverse()
print words
print '===='
keywords2 = []
i = 0
for tags in keywords_class:
keywords2.append([])
for word in wordList:
if word in tags:
keywords2[i].append((word,1))
else:
keywords2[i].append((word,0))
i+=1
g = kmeans.open_ubigraph_server()
result,clusters = kmeans.kcluster(g,lectures,keywords2,k=8)
dataClusters = []
i = 0
for cluster in result:
apCount = {}
for indice in cluster:
dados = keywords2[indice]
for word,count in dados:
apCount.setdefault(word,0)
apCount[word]+= count
words = apCount.items()
words.sort(key=operator.itemgetter(1))
words.reverse()
print words
print '===='
for user,wc in wordCounts.items():
socialNetworking[user] = {}
for word in wordlist:
socialNetworking[user].setdefault(word,0)
if word in wc:
socialNetworking[user][word] = wc[word]
items = socialNetworking.items()
users = [item[0] for item in items]
data = [item[1].items() for item in items]
#Step 05: Run a cluster algorithm (k-means)
g = kmeans.open_ubigraph_server()
#g = kmeans.open_ubigraph_server('http://IP:20738/RPC2')
result,clusters = kmeans.kcluster(g,users,data,k=15)
#Step 06: Presenting the results
usersResult = [[users[v] for v in result[i] ] for i in range(len(result))]
dataClusters = []
for cluster in usersResult:
apCount = {}
for user in cluster:
data = socialNetworking[user]
for word,wc in data.items():
apCount.setdefault(word,0)
apCount[word]+= wc
words = apCount.items()
words.sort(key=operator.itemgetter(1))