TC2011 - Bayes Network Lab

Setup

Before you begin, perform the following two steps:

Querying a Bayes Network

For this lab, you have to write a program to Query a Bayes Network

Objective

Your assignment is to build a program that reads the structure and probability tables of a Bayes Network, and then query it. Querying a Bayes network means to compute the probability of the query variables being true, given certain evidence.

Input/Output

Your program must read from stdin and output to stdout. See here for instructions how to do this.

Basically, your program will be called by executing run < [problem_file], where problem_file is a file describing the problem (see below). Your program should output the solution as text to the terminal.

Input

The input consists of three sections:

Nodes: a list of variable nodes in the Bayes Network, separated by a comma.
Probabilities: the probability tables associated variable nodes. Entries of a probability table have the following format:

VariableAssignment [| [VariableAssigment]*]? = Probability

This encodes the probability of a variable being true, given its parent nodes in the Bayes Network having a certain value.

A VariableAssignment is formatted as [+-]VariableName. When preceded by a +, it means that the variable is true. A minus (-) indicates the variable is false.
Queries: one or more queries. A query has the following format

[QueryAssignment]* [| [EvidenceAssigment]*]?

Both QueryAssignment and EvidenceAssigment are VariableAssigments.

A query basically requests your program to compute the probability of the query variables having a certain value, given the evidence variables having a certain value.

Output

Output consists multiple lines, one per query. For each query, output the probability, rounded to 7 digits.

Example

The following example encodes a Bayes network for a simple diagnostics test to determine whether a patient is ill.

The network consists of two nodes: "Ill" and "Test". There is one edge from "Ill" to "Test".

There are 3 queries defined.

Input:

[Nodes]
Test, Ill

[Probabilities]
+Ill = 0.001
+Test|+Ill=0.9
+Test|-Ill=0.05

[Queries]
+Ill
-Ill|+Test
+Test|+Ill

Output:

0.001
0.9823009
0.9

More examples can be found in the testcases directory

gomitaana/bayesian