Use of maximum entropy method as a methodology for probabilistic reasoning

Current methodologies which use probability theory for handling uncertainty are designed primarily for use in expert systems and require very specific knowledge for encapsulation. However, there are other situations where it would be desirable to use probabilistic reasoning but it would not be realistic to expect the knowledge to be available in a very specific form. For example, in some knowledge domains, such as engineering/manufacturing, knowledge exists which is not in the form required by these systems, and, for some purposes, such as decision support, it is quite possible for the knowledge provider to know about primary cause/effect relationships but not be in a position to assert that other relationships are nonexistent. The paper presents a method for reasoning in small knowledge domains which overcomes the above problem. The method invokes maximum entropy theory to estimate missing information and provide advice based on the knowledge available. The method is also shown to be capable of encapsulating whatever knowledge is available. The paper concludes by looking at the heavy computational requirements of the maximum entropy method for large knowledge domains, and it suggests that these might be alleviated, in many practical cases, because of the sparse nature of human knowledge in any domain. This will be the subject of further work.