Home

A04: Lazy Doctor

Using Problog (installed on londo using command problog), represent the diagnostic graph below. Make the probabilities unknown (“tunable”) and create examples of evidence to train the model. Then perform meaningful queries and write notes indicating their output. Further support the following constraints: pregnant can only be true for females, and if someone has a cold then they don’t have the flu, and vice versa. As usual, make a bitbucket repo and add me as a reader.

Grading

  • 5 pts: all requirements met
  • 4 pts: all requirements except constraints met
  • 3 pts: model built (perhaps missing constraints) but no evidence or learning takes place
  • 0-2 pts: partial completion of the above points

CSCI 431 material by Joshua Eckroth is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License. Source code for this website available at GitHub.