Investigating and extending P-log

This dissertation focuses on the investigation and improvement of knowledge representation language P-log that allows for both logical and probabilistic reasoning.
In particular, we extend P-log with new constructs to increase its expressive power and usability, clarify its semantics, define a new class of coherent (i.e., logically and probabilistically consistent) P-log programs and develop an inference algorithm for the programs from the new class. We also present the performance results of the preliminary implementation of the new algorithm. The results demonstrate that the new algorithm can substantially increase the performance of P-log inference on a number of important examples.