Intelligence analysis method and system using subjective logic
Method of and system for analysing a set of exhaustive and exclusive hypotheses, including assessing and assigning base rates for each hypothesis; determining a set of items of evidence that are relevant to, have a causal influence on, or would disconfirm more than one hypothesis; assessing and assigning base rates for each item of evidence; deciding, for each item of evidence, whether the item should be treated as being a causal influence or diagnostic indicator with respect to the set of the hypotheses; if the item of evidence is to be treated as a causal influence—making a judgement as to the likelihood of each hypothesis, both if the evidence were true, and also if the evidence were false; if the item of evidence is to be treated as a diagnostic indicator—making a judgement as to the evidence being true if the hypothesis were true; assessing the belief for each item of evidence being true; deciding a set of interim beliefs in each hypothesis for each individual item of evidence by: employing a conditional inference operator for evidence that is to be treated as a causal influence; and employing a reverse conditional inference operator for evidence that is to be treated as a diagnostic indicator; and deciding the overall belief in each hypothesis by employing a consensus operator on the respective set of interim beliefs.
1. A computer implemented method of analysis of competing hypotheses in estimative intelligence, said method comprising the steps of:
a. deciding on a plurality of possible hypotheses to be considered;
b. identifying significant items of evidence for and against each of said plurality of hypotheses;
c. configuring a processor to construct and store onto a memory a model for analyzing the hypotheses by:
i. producing a set of exhaustive and exclusive hypotheses, wherein only one hypothesis may be true;
ii. assessing and assigning base rates for each hypothesis;
iii. determining from said significant items of evidence identified in relation to respective hypotheses a set of items of evidence that are relevant to, have a causal influence on or would disconfirm more than one hypothesis;
iv. assessing and assigning base rates for each item of evidence;
v. deciding for each item of evidence whether the item should be treated as being a causal influence or diagnostic indicator with respect to the set of the hypotheses;
vi. if the item of evidence is to be treated as a causal influence, making a judgment as to the likelihood of each hypothesis:
A. if the evidence were true, and
B. if the evidence were false;
vii. if the item of evidence is to be treated as a diagnostic indicator, making a judgment as to the likelihood of the evidence being true:
A. if the hypothesis were true;
d. assessing the belief for each item of evidence being true;
e. deciding a set of interim beliefs in each hypothesis for each individual item of evidence by:
i. employing a conditional inference operator for evidence that is to be treated as a causal influence; and
ii. employing a reverse conditional inference operator for evidence that is to be treated as a diagnostic indicator;
f. deciding the overall belief in each hypothesis by employing a consensus operator on the respective set of interim beliefs; and
g. outputting a set of beliefs representing the certainty and likelihood of each hypothesis.
2. The analysis method of claim 1 wherein the step of identifying the set of significant items of evidence is not limited to what evidence is already known or believed to be discoverable.
3. The analysis method of claim 1 wherein the steps of assigning base rates to hypotheses and to the items of evidence involves assigning prior probabilities to each hypothesis and to each item of evidence, respectively.
4. The analysis method of claim 1 wherein competing hypotheses in the set are alternative explanations of the evidence.
5. The analysis method of claim 1 wherein the evidence includes possible causal influences and diagnostic indicators for one or more of the hypotheses.
6. The analysis method of claim 1 wherein the belief for an item of evidence being true includes both the certainty and likelihood of the item of evidence being true.
7. The analysis method of claim 6 wherein the likelihood of the belief an item of evidence or hypothesis being true expresses the probability of the item of evidence being true.
8. The analysis method of claim 6 wherein the certainty of the belief an item of evidence or hypothesis being true expresses the strength of the claim about the assigned likelihood being accurate.
9. The analysis method of claim 1 wherein deciding the belief in each interim hypothesis for each item of evidence is considered apart from the general set of evidence under consideration in relation to the hypothesis.
10. The analysis method of claim 1 wherein the model is comprised of a set of conditionals for all items of evidence and the hypotheses, wherein said model can be used to evaluate a complete or incomplete set of evidence by inputting a set of observable evidence together with expressions of vacuous beliefs where no data is available.
11. A system for analyzing a set of exhaustive and exclusive hypotheses in estimative intelligence, said system comprising:
at least one processor for executing instructions, a memory coupled to the processor, a data storage system for reading media having sequences of instructions stored thereon coupled to the data storage system, input/output for delivering data to and from the memory or to and from the data storage system and a user interface allowing for interaction with said instruction sequences by users, which sequences also causes said at least one processor to execute the steps of:
assessing and assigning base rates for each hypothesis;
determining a set of items of evidence that are relevant to, have a causal influence on or would disconfirm more than one hypothesis;
assessing and assigning base rates for each item of evidence;
deciding, for each item of evidence, whether the item should be treated as being a causal influence or diagnostic indicator with respect to the set of the hypotheses;
if the item of evidence is to be treated as a causal influence, making a judgment as to the likelihood of each hypothesis, both if the evidence were true, and also if the evidence were false;
if the item of evidence is to be treated as a diagnostic indicator, making a judgment as to the evidence being true if the hypothesis were true;
assessing the belief for each item of evidence being true;
deciding a set of interim beliefs in each hypothesis for each individual item of evidence by:
employing a conditional inference operator for evidence that is to be treated as a causal influence; and
employing a reverse conditional inference operator for evidence that is to be treated as a diagnostic indicator;
deciding the overall belief in each hypothesis by employing a consensus operator on the respective set of interim beliefs; and
outputting a set of beliefs representing the certainty and likelihood of each hypothesis.
12. The estimative intelligence analysis system of claim 11 wherein the step of identifying the set of significant items of evidence is not limited to what evidence is already known or believed to be discoverable.
13. The estimative intelligence analysis system of claim 11 wherein the steps of assigning base rates to hypotheses and to the items of evidence involves assigning prior probabilities to each hypothesis and to each item of evidence, respectively.
14. The estimative intelligence analysis system of claim 11 wherein competing hypotheses in the set are alternative explanations of the evidence.
15. The estimative intelligence analysis system of claim 11 wherein the evidence includes possible causal influences and diagnostic indicators for one or more of the hypotheses.
16. The estimative intelligence analysis system of claim 11 wherein the belief for an item of evidence being true includes both the certainty and likelihood of the item of evidence being true.
17. The estimative intelligence analysis system of claim 16 wherein the likelihood of the belief an item of evidence or hypothesis being true expresses the probability of the item of evidence being true.
18. The estimative intelligence analysis system of claim 16 wherein the certainty of the belief an item of evidence or hypothesis being true expresses the strength of the claim about the assigned likelihood being accurate.
19. The estimative intelligence analysis system of claim 11 wherein deciding the belief in each interim hypothesis for each item of evidence is considered apart from the general set of evidence under consideration in relation to the hypothesis.
20. The estimative intelligence analysis system of claim 11 wherein the model is comprised of a set of conditionals for all items of evidence and the hypotheses, wherein said model can be used to evaluate a complete or incomplete set of evidence by inputting a set of observable evidence together with expressions of vacuous beliefs where no data is available.
21. A computer program product for analysis of competing hypotheses in estimative intelligence, the computer program product comprising:
computer-readable storage media having computer-readable sequences of instructions stored on the computer-readable storage media, wherein the sequences of instructions comprise:
first instructions configured to decide on a plurality of possible hypotheses to be considered;
second instructions configured to identify significant items of evidence for and against each of said plurality of hypotheses;
third instructions configured to construct a model for analyzing the hypotheses by:
i. producing a set of exhaustive and exclusive hypotheses, wherein only one hypothesis may be true;
ii. assessing and assigning base rates for each hypothesis;
iii. determining from said significant items of evidence identified in relation to respective hypotheses a set of items of evidence that are relevant to, have a causal influence on or would disconfirm more than one hypothesis;
iv. assessing and assigning base rates for each item of evidence;
v. deciding for each item of evidence whether the item should be treated as being a causal influence or diagnostic indicator with respect to the set of the hypotheses;
vi. if the item of evidence is to be treated as a causal influence, making a judgment as to the likelihood of each hypothesis:
A. if the evidence were true, and
B. if the evidence were false; and
vii. if the item of evidence is to be treated as a diagnostic indicator, making a judgment as to the likelihood of the evidence being true:
A. if the hypothesis were true;
fourth instructions configured to access the belief for each item of evidence being true;
fifth instructions configured to decide a set of interim beliefs in each hypothesis for each individual item of evidence by:
i. employing a conditional inference operator for evidence that is to be treated as a causal influence; and
ii. employing a reverse conditional inference operator for evidence that is to be treated as a diagnostic indicator;
sixth instructions configured to decide the overall belief in each hypothesis by employing a consensus operator on the respective set of interim beliefs; and
seventh instructions configured to output a set of beliefs representing the certainty and likelihood of each hypothesis.