Download presentation

Presentation is loading. Please wait.

Published byJalynn Suter Modified over 2 years ago

1
Fundamentals of Dempster-Shafer Theory presented by Zbigniew W. Ras University of North Carolina, Charlotte, NC College of Computing and Informatics University of North Carolina, Charlotte www.kdd.uncc.edu

2
Dempster-Shafer Theory based on the idea of placing a number between zero and one to indicate the degree of belief of evidence for a proposition.

3
Basic Probability Assignment - function m: 2^X [0,1] such that: (1) m( )=0, (2) [m(Y) : Y X] = 1 /total belief/. m(Y) – basic probability number of Y. Belief function over X - function Bel: 2^X [0,1] such that: Bel(Y)= [m(Z): Z Y]. FACT 1: Function Bel: 2^X [0,1] is a belief function iff (1)Bel( )=0, (2) Bel(X)= 1, (3) Bel( {A(i): i {1,2,…,n}) = [(-1)^{|J|+1} Bel( {A(i): i J}) : J {1,2,…,n}] for every positive integer n and all subsets A(1), A(2), …, A(n) of X FACT 2: Basic probability assignment can be computed from: m(Y) = [ (-1)^{|Y – Z| Bel(Z): Z Y], where Y X.

4
Example: basic probability assignment Xabcd x1x1 0L x2x2 0SL x3x3 P1L x4x4 3R1L x5x5 22L x6x6 P2L x7x7 3P2H m_a({x1,x2,x3,x6})=[2+2/3]/7=8/21 m_a({x3,x6,x5})=[1+2/3]/7=5/21 m_a({x3,x6,x4,x7})=[2+2/3]/7=8/21 Basic probability assignment (given) m({x1,x2,x3,x6})=8/21 m({x3,x6,x5})=5/21 m({x3,x6,x4,x7})=8/21 defines attribute m_a 1) m_a uniquely defined for x1,x2,x4,x5,x7. 2)m_a undefined for x3,x6. m_a(x1)=m_a(x2) =a1, m_a(x5)=a2,…..

5
Example: basic probability assignment Basic probability assignment – m: X={x1,x2,x3,x4,x5} m(x1,x2,x3)=1/2, m(x1,x2)=1/4, m(x2,x4)=1/4 Belief function: Bel({x1,x2,x3,x5})= ½ + ¼ = ¾, ……….. Focal Element and Core Y X is called focal element iff m(Y) > 0. Core – the union of all focal elements. Doubt Function - Dou: 2^X [0,1], Y X Dou(Y) = Bel( Y). Plausibility Function – Pl(Y) = 1 – Dou(Y) Pl(Y)= [m(Z): Z Y ]

6
{1,2} 1/4 {1,2} {2,3} 3/4 {1,2} {1,3} 1/2 {1,2} {1} 0 {1,2} {2} 0 {1,2} {3} 1/2 m({3})=1/2, m({2,3})=1/4, m({1,2})=1/4. Pl({1,2}) = m({2,3})+m({1,2}) = ½, Pl({1,3})= m({3})+m({2,3})+m({1,2}) = 1 Core={1,2,3}

7
Properties: - Bel( ) = Pl( ) = 0 - Bel(X) = Pl(X) = 1 - Bel(Y) Pl(Y) - Bel(Y) + Bel( Y) 1 - Pl(Y) + Pl( Y) 1 - if Y Z, then Bel(Y) Bel(Z) and Pl(Y) Pl(Z) Bel: 2^X [0,1] is called a Bayesian Belief Function iff 1)Bel( ) = 0 2)Bel(X) = 1 3)Bel (Y Z)= Bel(Y) + Bel(Z), where Y, Z X, Y Z = Fact: Any Bayesian belief function is a belief function.

8
The following conditions are equivalent: 1)Bel is Bayesian 2)All focal elements of Bel are singletons 3)Bel = Pl 4)Bel(Y) + Bel( Y) = 1 for all Y X

9
Dempster’s Rule of Combination Bel1, Bel2 – belief functions representing two different pieces of evidence which are independent. Domain = {x1,x2,x3} Bel1 Bel2 – their orthogonal sum /Dempster’s rule of comb./ m1, m2 – basic probability assignments linked with Bel1, Bel2. {x1,x2} 1/4 {x1,x2,x3} 3/2 {x2,x4} 1/4 {x2} 3/8 {x2} 3/32 {x2} 3/16 {x2} 3/32 {x1,x2,x4} 3/8 {x1,x2} 3/32 {x1,x2} 3/16 {x2,x4} 3/32 {x1,x2,x3} 1/4 {x1,x2} 1/16 {x1,x2,x3} 1/8 {x2} 1/16 m1 m2 (m1 m2)({x1,x2})=3/32+3/16+1/16=11/32 (m1 m2)({x1,x2,x3})=1/8 (m1 m2)({x2})=3/32+3/16+3/32+1/16=7/16 (m1 m2)({x2,x4})=3/32

10
Thank You Questions?

Similar presentations

OK

Copyright © 2011 Pearson Education, Inc. Combinations, Labeling, and the Binomial Theorem Section 8.5 Sequences, Series, and Probability.

Copyright © 2011 Pearson Education, Inc. Combinations, Labeling, and the Binomial Theorem Section 8.5 Sequences, Series, and Probability.

© 2017 SlidePlayer.com Inc.

All rights reserved.

Ads by Google

Ppt on game theory band Ppt on eia report bakken Ppt on wireless intelligent network Ppt on limitation act ontario Ppt on p&g products brands tide Ppt on natural resources for class 8 Ppt on heredity and evolution class x Ppt on carburetor repair Ppt on motion for class 9 free download Safety ppt on working at height