An Initiative Learning Algorithm Based on System Uncertainty

Initiative-learning algorithms are characterized by and hence advantageousfor their independence of prior domain knowledge. Usually, their induced results could moreobjectively express the potential characteristics and patterns of information systems.Initiative-learning processes can be effectively conducted by system uncertainty, becauseuncertainty is an intrinsic common feature of and also an essential link between information systemsand their induced results. Obviously, the effectiveness of such initiative-learning framework isheavily dependent on the accuracy of system uncertainty measurements. Herein, a more reasonablemethod for measuring system uncertainty is developed based on rough set theory and the conception ofinformation entropy; then a new algorithm is developed on the bases of the new system uncertaintymeasurement and the Skowron's algorithm for mining prepositional default decision rules. Theproposed algorithm is typically initiative-learning. It is well adaptable to system uncertainty. Asshown by simulation experiments, its comprehensive performances are much better than those ofcongeneric algorithms.