Machine Learning is a branch of the artificial intelligence research. It has attracted a lot of attention since the mid-1950's. Machine Learning focuses on the induction of knowledge and the refinement of skill. A learning machine is a computer program that has the power to mimic the intelligence of human, both in the learning and the problem solving processes. The project, A Learning Program (ALPS), has been initiated to develop a program capable of learning quantitative knowledge, and applying them to solve problems.
This thesis, the problem solving subsystem, is a part of the learning system that concentrates on the problem solving process. The system is designed to solve problems that involves relationships which satisfy the transitive property, specifically, three types of problems are addressed: "problem is", "problem which", and "problem what". An interface is provided to demonstrate that the all the problems are solved correctly based on the given set of input information.
The design of the system is done using object oriented paradigms, and expressed in UML notations. The system has been implemented using C++. The system also performs consistency checks for problem solving.