A Victim-Based Framework for Telecom Fraud Analysis: A Bayesian Network Model

Table 5

Bayesian network node table description.

Case	Bayesian nodes	Description
A 1	(A) Marriage	Marriage = married
A 2	(A) Marriage	Marriage = unmarried
B 1	(B) Sex	Sex = male
B 2	(B) Sex	Sex = female
C 1	(C) Age	Age = young
C 2	(C) Age	Age = middle age
C 3	(C) Age	Age = old age
D 1	(D) Work	Work = company
D 2	(D) Work	Work = school
D 3	(D) Work	Work = selfemployed person
D 4	(D) Work	Work = government
E 1	(E) Knowledge	Knowledge = high
E 2	(E) Knowledge	Knowledge = low
F 1	(F) Suspect during cheat	Suspect during cheat = yes
F 2	(F) Suspect during cheat	Suspect during cheat = no
G 1	(G) Cheat type	Cheat type = identity fraud
G 2	(G) Cheat type	Cheat type = shopping fraud
G 3	(G) Cheat type	Cheat type = inducement fraud
G 4	(G) Cheat type	Cheat type = fictional dangerous situation fraud
G 5	(G) Cheat type	Cheat type = daily consumption fraud
G 6	(G) Cheat type	Cheat type = phishing and Trojan virus fraud
G 7	(G) Cheat type	Cheat type = other types of fraud
H 1	(H) Communicate type	Communicate type = phone
H 2	(H) Communicate type	Communicate type = message
H 3	(H) Communicate type	Communicate type = social software
I 1	(I) Call the police	Call the police = yes
I 2	(I) Call the police	Call the police = no
J 1	(J) Property loss	Property loss = 0–1000
J 2	(J) Property loss	Property loss = 1000–5000
J 3	(J) Property loss	Property loss = 5000–20000
J 4	(J) Property loss	Property loss = 20000–50000
J 5	(J) Property loss	Property loss = 50000+