2013-08-30

Fallacyに対峙する

Avoiding Probabilistic Reasoning Fallacies in Legal Practice using Bayesian Networks
抜粋
"Indeed, we believe that for many people – and this includes from our own experience highly intelligent barristers, judges and surgeons, any attempt to use Bayes theorem to explain a fallacy is completely hopeless"
"The introduction of Bayes' theorem into a criminal trial plunges the jury into inappropriate and unnecessary realms of theory and complexity deflecting them from their proper task"
"The task of the jury is 'to evaluate evidence and reach a conclusion not by means of a formula, mathematical or otherwise, but by the joint application of their individual common sense and knowledge of the world to the evidence before them'"
"The introduction of long division into a criminal trial plunges the jury into inappropriate and unnecessary realms of theory and complexity deflecting them from their proper task"
"The task of the jury is 'to evaluate evidence and reach a conclusion not by means of a formula, mathematical or otherwise, but by the joint application of their individual common sense and knowledge of the world to the evidence before them'"
ベイジアンネットワークへ、そしてそれを「視覚表現」して法廷へ
- (1) Most people can understand and do it from scratch in very simple cases
- (2) Scientists have developed algorithms for doing it in the general case
- (3) There are computers that implement the algorithms to acceptable degrees of accuracy
"What we have demonstrated to you is how we revise our prior assumption when we observe a single piece of evidence. Although we were able to explain this to you from scratch, there is a standard calculation engine (accepted and validated by the mathematical and statistical community) which will do this calculation for us without having to go through all the details. In fact, when there is more than a single piece of evidence to consider it is too timeconsuming and complex to do the calculations by hand, but the calculation engine will do it instantly for us. This is much like relying on a calculator to do long division for us. You do not have to worry about the accuracy of the calculations; these are guaranteed. All you have to worry about is whether our original assumptions are easonable. But we can show you the results with a range of different assumptions"

strict digraph G{
	graph [compound=true];
	graph [charset="UTF-8",fontname="MS UI Gothic"];
	node[charset="UTF-8",fontname="MS UI Gothic"];
	edge[charset="UTF-8",fontname="MS UI Gothic"];

	subgraph cluster_persons {
		node [style=filled,color="#FFA500"];
		p1 p2 p3;
		label = "人";
	}
	subgraph cluster_true_genotype {
		node [style=filled,color="#EE82EE"];
		G1 G2 G3;
		label = "真のジェノタイプ";
	}
	subgraph cluster_obs_genotype {
		node [style=filled,color="#8B008B"];
		g1[label="X:調べたジェノタイプ"] g2 g3;
		label = "観測ジェノタイプ";

	}
	subgraph cluster_localized {
		node [style=filled,color="#00BFFF"];
		L1 L2 L3;
		label = "現場に存在";

	}
	subgraph cluster_resid {
		node [style=filled,color="#4169E1"];
		R1 R2 R3;
		label = "現場に残す";

	}
	subgraph cluster_motivation {
		node [style=filled,color="#A52A2A"];
		M1 M2 M3;
		label = "動機";

	}
	p1 -> G1
	p2 -> G2
	p3 -> G3
	G1 -> g1
	G2 -> g2
	G3 -> g3
	p1 -> L1
	p2 -> L2
	p3 -> L3
	L1 -> R1
	L2 -> R2
	L3 -> R3
	p1 -> M1
	p2 -> M2
	p3 -> M3
	L1 -> C
	L2 -> C
	L3 -> C
	M1 -> C
	M2 -> C
	M3 -> C
	G1 -> F
	G2 -> F
	G3 -> F
	R1 -> F
	R2 -> F
	R3 -> F
	/*
	p1 -> G1 [ltail=cluster_persons, lhead=cluster_true_genotype];
	G1 -> g1 [ltail=cluster_true_genotype, lhead=cluster_obs_genotype];
	p1 -> L1 [ltail=cluster_persons, lhead=cluster_localized];
	L1 -> R1 [ltail=cluster_localized, lhead=cluster_resid];
	p1 -> M1 [ltail=cluster_persons, lhead=cluster_motivation];
	L1 -> C [ltail=cluster_localized];
	M1 -> C [ltail=cluster_motivation];
	*/
	F[label="遺留物の真のジェノタイプ",style=filled,color="#bbbb22"];
	f[label="Y=X?:遺留物の観察ジェノタイプ",style=filled,color="#ffbb22"];
	G1 -> F [ltail=cluster_true_genotype];
	R1 -> F [ltail=cluster_resid];
	F -> f;
	C[label="犯人",style=filled,color="#ff5555"]
}

dot -Kdot -Tsvg C:\Users\ryamada\Desktop\MybookPDF\文書管理\bn犯人.txt -o C:\Users\ryamada\Desktop\MybookPDF\文書管理\bn犯人.svg

dot -Kdot -Tpng C:\Users\ryamada\Desktop\MybookPDF\文書管理\bn犯人.txt -o C:\Users\ryamada\Desktop\MybookPDF\文書管理\bn犯人.png