EVIDENCE DIRECTED GENERATION OF PLAUSIBLE CRIME SCENARIOS WITH IDENTITY RESOLUTION

Figures & data

FIGURE 1 The fuzzy quantity space of Chance.

FIGURE 2 Outline of the approach.

Table

Algorithm 1 Backward propagation ([ztilde] c , [ztilde] r )
1:	Re([ztilde]antecedent) ← 
2:	Im([ztilde]antecedent) ← 
3:	S Re  = 0
4:	S Im  = 0
5:	for j = 1: n do
6:	if S(Re([ztilde] c ), V FN j  × [ztilde] r ) > S Re then
7:	S Re  = S(Re([ztilde] c ), V FN j  × [ztilde] r )
8:	Re([ztilde]antecedent) = V FN j
9:	end if
10:	if S(Im([ztilde] c ), V FN j  × [ztilde] r ) > S Im then
11:	S Im  = S(Im([ztilde] c ), V FN j  × [ztilde] r )
12:	Im([ztilde]antecedent) = V FN j
13:	end if
14:	end for
15:	return [ztilde]antecedent

Table

Algorithm 2 Forward propagation ([ztilde] p 1 ,…, [ztilde] p n , [ztilde] r )
1:	Calculate [ztilde]antecedent
2:	[ztilde]new = [ztilde]antecedent × [ztilde] r
3:	S Re  = 0
4:	S Im  = 0
5:	for j = 1: n do
6:	if S(V FN j , Re([ztilde]new)) > S Re then
7:	S Re  = S(V FN j , Re([ztilde]new))
8:	Re([ztilde] c ) = V FN j
9:	end if
10:	if S(V FN j , Im([ztilde]new)) > S Im then
11:	S Im  = S(V FN j , Im([ztilde]new))
12:	Im([ztilde] c ) = V FN j
13:	end if
14:	end for
15:	return [ztilde] c

FIGURE 3 A social network with Connected-Triple.

FIGURE 4 Definition of fuzzy sets for describing a similarity measure.

TABLE 1 Results of Backward Chaining Due to e ₁

Rule index	Re([ztilde]antecedent)	Im([ztilde]antecedent)
r 1	1	1
r 2	N/A	0

FIGURE 5 Generated scenario space.

FIGURE 6 A link network with fuzzy linguistic weights.

TABLE 2 Results of Forward-Chaining Propagation Within MF ₄

Rule index	Re([ztilde] c )	Im([ztilde] c )
r 1	H	H
r 2	VL	VL
r 3	VL	VL
r 4	H	L

FIGURE 7 Updated scenario space.