-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathnodeBox
168 lines (116 loc) · 6.18 KB
/
nodeBox
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
import nodeCluster
import numpy as np
import random
# nodeBox: collecting all node with it neighbor index which is in clusterNode structure
class nodeBox(object):
# list for the clusterNode: input of the clustering
box = []
# list for the clusterName: cluster name and its member which gives output of the clustering
clusterList = []
# recording the node which has been visited ....for the algorithms
visitedList = []
def ranSelection(self, rowNum, num):
if num <= rowNum:
indexSample = random.sample(range(rowNum), num)
else:
print("Pick a num <= %" % (n))
return indexSample
def __init__(self, D, n, num, eps):
for i in range(n):
# for each node, first get randomly a number of nodes(not including itself) to see if they are its neighbor
listSample = self.ranSelection(n, num)
if i in listSample:
listSample.remove(i)
# put it into nodeCluster
tempNode = nodeCluster.nodeCluster(i, listSample)
# use getTempNeighor methods to get node list in its neighbor based on eps
currentNode = nodeCluster.nodeCluster(i,tempNode.getTempNeighbor(D, eps))
# put it into nodeBox
self.box.append(currentNode)
#self.box.append(tempNode)
def getNode(self, i):
return self.box[i]
def getClusterNameList(self):
clusterNameList = []
for cluster in self.clusterList:
clusterNameList.append(cluster.name)
return clusterNameList
# method for clustering
def clusterDB(self, rowNum, minNeigh):
#print(len(self.box))
# firstly put noiseclustering into nodeBox.clusterList
clusterNoise=clusterName(-1, [])
self.clusterList.append(clusterNoise)
# checking all nodes in nodeBox.box
for currentNode in self.box:
if currentNode.indexCurrentNode not in self.visitedList:
#print(currentNode.indexCurrentNode)
#print(self.visitedList)
# if not in visitedList then generate an new cluster using clusterName
clusterTemp = clusterName(currentNode.indexCurrentNode, [])
#print(currentNode.listSample)
#print(clusterTemp.name)
#print("pass")
# add current node's neighbor in cluster's member (clusterName.member)
# also record these nodes to be visited
clusterTemp.addMember(currentNode.listSample)
self.visitedList.append(currentNode.indexCurrentNode)
#print(clusterTemp.getMember())
#print(currentNode.indexCurrentNode)
# put current node into oldSet and itself and its neighbor into newSet
newSet = set([currentNode.indexCurrentNode]) | set(clusterTemp.getMember())
oldSet = set([currentNode.indexCurrentNode])
# checking the nodes whose neighbour has node in newSet
# if so put this node into newset and update visitedlist
for i in [j for j in range(rowNum) if j not in list(oldSet)]:
currentTempNode =self.getNode(i)
intersectSet = set(currentTempNode.listSample).intersection(oldSet)
if len(intersectSet) != 0:
self.visitedList.append(currentTempNode.indexCurrentNode)
newSet= newSet | set([currentTempNode.indexCurrentNode])
# as long has difference keep puting newnodes and its neighbor into this cluster's member
while len(newSet-oldSet) != 0:
diff =newSet-oldSet
diff = list(diff)
#print(diff)
#visitedList.append(diff)
#print(clusterTemp.getMember())
clusterTemp.addMember(diff)
oldSet = newSet
#newSet = oldSet
for i in diff:
currentNeighborNode =self.getNode(i)
self.visitedList.append(currentNeighborNode.indexCurrentNode)
newSet = newSet | set(currentNeighborNode.listSample) | set([currentNeighborNode.indexCurrentNode])
# checking the nodes whose neighbour has node in newSet
# if so put this node into newset and update visitedlist
for i in [j for j in range(rowNum) if j not in list(oldSet)]:
currentTempNode =self.getNode(i)
intersectSet = set(currentTempNode.listSample).intersection(oldSet)
if len(intersectSet) != 0:
self.visitedList.append(currentTempNode.indexCurrentNode)
#clusterTemp.appMember(currentTempNode.indexCurrentNode)
newSet= newSet | set([currentTempNode.indexCurrentNode])
#print(len(clusterTemp.getMember()))
# check the member size for each cluster and get rid of those whose member size < minNeigh
if len(clusterTemp.getMember())<= minNeigh:
tempList = clusterTemp.getMember()
self.clusterList[0].addMember(tempList)
else:
self.clusterList.append(clusterTemp)
#print(len(clusterTemp.getMember()))
#print("pass")
#print(clusterTemp.getMember())
class clusterName(object):
member = []
def __init__(self, name, member):
self.name = name
self.member = member
def addMember(self, listNeigh):
self.member = list(set(listNeigh) | set(self.member))
def getMember(self):
return self.member
def appMember(self, currentNode):
self.member.append(currentNode)
def nameModify(self, newName):
self.name = newName