Difference between revisions of "ISSS608 2017-18 T3 Assign Vigneshwar Ramachandran Vadivel Q2"

From Visual Analytics and Applications
Jump to navigation Jump to search
Line 39: Line 39:
  
 
Combine the four data sources for group that the insider has identified as being suspicious and locate the group in the larger dataset. Determine if anyone else appears to be closely associated with this group. Highlight which employees are making suspicious purchases, according to the insider’s data.
 
Combine the four data sources for group that the insider has identified as being suspicious and locate the group in the larger dataset. Determine if anyone else appears to be closely associated with this group. Highlight which employees are making suspicious purchases, according to the insider’s data.
 +
 +
To infer the big picture of the suspicious network it is necessary to identify the entire network of those employees.
 +
We can prepare the data by filtering only the persons of interest in the source and target of the communication transaction. Once we have the network of employee , we can locate the group in the larger dataset.
 +
 +
With the help of Gephi, we can try to map the network of communications among the larger dataset, by converting the data into edges and nodes.
 +
Using the layout Force Atlas 2 a network spatialization algorithm to differentiate the communication pattern within this network.
 +
 +
This visualization of larger dataset clearly shows that only few actors are connecting the entire network. All of the connecting nodes are our suspicious employee who plays an important role in influencing the communication among employee in the organization.
 +
 +
Richard fox, Meryl Pastuch , Tobi Gatlin and Lizabeth Jindra acts as connecting node for a large group of employee. This can be considered as an indicator about the influence of these employee on the entire network.
 +
 +
Degree centrality also helps us to understand the flow of communication from each employee in the entire network. We can refer the indegree and out degree measures to assess the importance of a node for catching whatever is flowing through the network
 +
 +
Clearly Richard fox acts as an influencer in the entire network by the degree centrality measures.

Revision as of 16:55, 4 July 2018

MC3 Banner.png

Intro

Approach

Findings

Conclusion

Question 1

Question 2

Question 3

Question 4

Combine the four data sources for group that the insider has identified as being suspicious and locate the group in the larger dataset. Determine if anyone else appears to be closely associated with this group. Highlight which employees are making suspicious purchases, according to the insider’s data.

To infer the big picture of the suspicious network it is necessary to identify the entire network of those employees. We can prepare the data by filtering only the persons of interest in the source and target of the communication transaction. Once we have the network of employee , we can locate the group in the larger dataset.

With the help of Gephi, we can try to map the network of communications among the larger dataset, by converting the data into edges and nodes. Using the layout Force Atlas 2 a network spatialization algorithm to differentiate the communication pattern within this network.

This visualization of larger dataset clearly shows that only few actors are connecting the entire network. All of the connecting nodes are our suspicious employee who plays an important role in influencing the communication among employee in the organization.

Richard fox, Meryl Pastuch , Tobi Gatlin and Lizabeth Jindra acts as connecting node for a large group of employee. This can be considered as an indicator about the influence of these employee on the entire network.

Degree centrality also helps us to understand the flow of communication from each employee in the entire network. We can refer the indegree and out degree measures to assess the importance of a node for catching whatever is flowing through the network

Clearly Richard fox acts as an influencer in the entire network by the degree centrality measures.