MINDLAB University of Maryland,College Park
VAST 2009 Challenge
Challenge 2 - Social Network and Geospatial
Authors and Affiliations:
Neha Gupta neha@cs.umd.edu PRIMARY CONTACT
Christian Almazan almazan@cs.umd.edu
Dr.Ashok Agrawala agrawala@cs.umd.edu
Tool(s):
Visual Links by Visual Analytics Inc. (http://www.visualanalytics.com/)
This tool is a product developed around 10 years back by Visual Analytics a small company in Frederick,MD. It is mainly used by Department of Police and Security Departments to detect frauds and illegal activities.
Video:
ANSWERS:
MC2.1: Which of the two social structures, A or B, most closely match the scenario you have identified in the data?
A
MC2.2: Provide the social network structure you have identified as a tab delimitated file. It should contain the employee, one or more handler, any middle folks, and the localized leader with their international contacts. What are the Flitter names of the persons involved? Please identify only key connections (not all single links for example) as well as any other nodes related to the scenario (if any) you may have discovered that were not described in the two scenarios A and B above.
MC2.3: Characterize the difference between your social network and the closest social structure you selected (A or B). If you include extra nodes please explain how they fit in to your scenario or analysis.
We chose the network A for the problem since B is not possible by the description as if we have three middleman and the middleman can talk to one or two others in the organization then each middleman should have 2-3 contacts. But in the data no person entity hs 2-3 contacts, hence scenario B is not possible.
According to scenario A,
· The employee has about 40 flitter contacts.
· He is linked to 3 handlers who each have about 30-40 contacts and they don’t talk to each other.
· They all speak to one common person who is called middleman “Boris”. Boris communicates with at most two other men in the organization so total number of contacts Boris can have are 4-5.
· Boris talks to a guy who is the Fearless leader containing more than 100 contacts and several international contacts.
We performed a draw search query using our tool, we searched for all the people who have between 35-45 contacts and are connected to the people who have 30-40 contacts.We came up with few possible combinations. We went ahead and further expanded the network in the tool to find out the handlers who do not talk to each other and talk to atleast one common person (middleman). The most probable network that fitted was below (the whole suspect organization is given is Flitter.txt) :
· employee 100
· 194,261,563 as handlers
· 4994 as middleman
· 4 Fearless Leader
Figure 1:Core Suspicious Network
.
The difference between our social network and A is that our middleman is not only related to the fearless leader and handlers but also another person #1612. We are not sure if #1612 is part of the organization as mentioned in the scenario A
Another interesting aspect is that 1612 is also connected to a friend of fearless leader with id# 3651.
Figure 2:#1612 Other suspect
There are few other people who are related to both the employee and the fearless leader. As shown in the figure below IDs # 351, 14, 170 and22 are connected to both ID#4 and ID#100.
Nothing is mentioned about such people in scenario A so we put them in our related other list. They can possibly be part of the suspect organization.
Figure 3:Links between fearless leader and employee
Another few suspicious people are the ones connected to the fearless leader and the handlers #563 and #261. Below figure shows connections between fearless leader and handler #261. They are ID # 584, 4206, 1, 250.
. Similarly there are people related to both handler #563 and fearless leader. They are #1975, 102,1834, 686
Figure 4:Links between handler ID#563 and Fearless Leader
There are no common links between handler 194 and fearless leader 4.
Fearless leader has several international contacts in across countries of Posana,Transak and Trium and has over 256 connections as shown in the figure below.
Figure 5: Leader with international contacts
MC2.4: How is your hypothesis about the social structure in Part 1 supported by the city locations of Flovania? What part(s), if any, did the role of geographical information play in the social network of part one?
· The employee and the three handlers all reside in the same city which may make it easy for the handlers to communicate and meet with the employees.
· The fearless leader is in Kouvnic, Flovania which is a medium sized city.However, Kouvnic has a large number of flitter links compared to other mid-sized cities (related to 2.5).
· The middleman is also not in a city of anyone else, in fact, the middleman doesn't reside in a city of anyone else in the Flitter.txt table which we have provided .Since he is an articulation point in the network, he is an important person and he might want to be in a safe haven.
· The fearless leader is connected to people in several cities and may have few direct people of interest - 14,170,22 and 14.(linked to employee) and 3651 (linked to middleman's related).
· The city with most contacts in the suspect organization is Prounov.
· Out of all people in the in the criminal network( provided in the Flitter.txt) , no one is in Pasko, Solank, Solvenz, or Ryzkland
· Most of the activity seems to be focused around Koul, Otello and Prounov.
· There is little activity in Sresk and Kannvic.
Figure 6: This figure presents the core entities.Each of the entities are colored according to city.
Below is
the figure of the suspect organization grouped by cities.
The data
below shows the count per city:
1)
Figure 7: Suspect network with cities
MC2.5: In general, how are the Flitter users dispersed throughout the cities of this challenge? Which of the surrounding countries may have ties to this criminal operation? Why might some be of more significant concern than others?
- The below data provides a
summary of the links between the cities and the flitter users. They are in
the order of city, country and count per city. Koul tops with the city
with most users.
1. Koul, Flovania, 1998
2. Solvenz, Flovania, 210
3. Kouvnic, Flovania, 798
4. Ryzkland, Flovania, 142
5. Otello, Posana, 147
6. Transpasko, Transak, 126
7. Pasko, Flovania, 147
8. Solank, Flovania, 135
9. Prounov, Flovania, 1707
10. Sresk, Flovania, 147
11. Tulamuk, Trium, 123
12. Kannvic, Flovania, 320
· There are 13 international contacts in our suspect list which are shown below in the order of count. The data is arranged as count of suspects in city, city, country.
1. 7, Otello, Posana
2. 3, Tulamuk, Trium
3. 3, Transpasko, Trium
· The most suspect able international country is Posana. But there are few contacts ~3 in Trium and Transak as well so they cannot be ruled out .
Figure 8: Suspect Organization grouped by country