3.3. NETWORK-BASED DETECTION 49
            
       C
0
    
   W        
          
             
           
               
           
            
               

Fake
Real
a
c

c
c
c
c
c
c
c

c
c
c
√
?
GC GC FC FC SMMP
            
        D    D 
  D    D  
3.3.6 HIERARCHICAL PROPAGATION NETWORK MODELING
              
          
           
                 
               
               
           
          
              
          
            
50 3. HOW SOCIAL CONTEXT HELPS
a
c

c
c
c
c
c
c
c

c

c
c
c
c
c
c
c
c

c
c
a
c
c
c
c
c
c
c
Macro-Level
M
icr
o
-
Level
              
   macro-level  micro-level      
            
     
Macro-Level Propagation Network     
           
            
            
  
Structural analysis          
               
              
               
          
.S
1
/ Tree depth            
       
.S
2
/ Number of nodes           
              
.S
3
/ Maximum Outdegree        
        
3.3. NETWORK-BASED DETECTION 51
.S
4
/ Number of cascades          
.S
5
/ Depth of node with maximum outdegree        
               
              
.S
6
/ Number of cascades with retweets        
   
.S
7
/ Fraction of cascades with retweets        
   
.S
8
/ Number of bot users retweeting         
    
.S
9
/ Fraction of bot users retweeting            
             
         
Temporal analysis           
             
              
              
           
            
             
.T
1
/ Average time difference between the adjacent retweet nodes     
      
.T
2
/ Time difference between the first tweet and the last retweets      
   
.T
3
/ Time difference between the first tweet and the tweet with maximum outdegree 
           
              

.T
4
/ Time difference between the first and last tweet posting news    
         
.T
5
/ Time difference between the tweet posting news and last retweet node in deepest cascade
           
              
         
52 3. HOW SOCIAL CONTEXT HELPS
.T
6
/ Average time difference between the adjacent retweet nodes in the deepest cascade 
           
.T
7
/ Average time between tweets posting news      
       
.T
8
/ Average time difference between the tweet post time and the first retweet time  
              
      
Micro-Level Propagation Network      
               
             
       
Structure analysis          
              
S
10
Tree depth            
       
S
11
Number of nodes          
               
  
S
12
Maximum Outdegree        
            
S
13
Number of cascade with with micro-level networks     
       
S
14
Fraction of cascades with micro-level networks       
         
Temporal analysis         
             
               
     
T
9
Average time difference between adjacent replies in cascade    
    
T
10
Time difference between the first tweet posting news and first reply node   
           
3.3. NETWORK-BASED DETECTION 53
T
11
Time difference between the first tweet posting news and last reply node in micro network
           
 
T
12
Average time difference between replies in the deepest cascade    
        
T
13
Time difference between first tweet posting news and last reply node in the deepest cas-
cade               

Linguistic analysis          
            
                  
             
           
             
             
             
L
1
Sentiment ratio          
              
              

L
2
Average sentiment          
              
      
L
3
Average sentiment of first level replies       
           
L
4
Average sentiment of replies in deepest cascade     
            
              
   
L
5
Sentiment of first level reply in the deepest cascade     
            
             
             
              
             
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset