A data-driven knowledge acquisition method based on system uncertainty

Data-driven knowledge acquiring approach is characterized by and hence advantageous for its unnecessity of prior domain knowledge. Therefore, it is very possible for its induced results to express the potential characteristics and patterns of decision information systems more objectively. Data-driven knowledge acquiring process can be effectively conducted by system uncertainty since uncertainty is an intrinsic common feature of and an essential link between decision information systems and their induced rule-like knowledge systems. Obviously, the effectiveness of such a data-driven knowledge acquiring framework depends heavily on whether system uncertainty can be measured reasonably and precisely. To find a suitable measure method for system uncertainty, various uncertainty measurements based on rough set theory are studied. Their algebraic characteristics and quantitative relations are disclosed. Their performances are comprehensively studied and compared. Then, a new data-driven knowledge acquiring algorithm is developed based on the optimal method for measuring system uncertainty and the algorithm of Professor Skowron for mining default decision rules. Results of simulation experiments illustrate that the proposed algorithm obviously outperforms other similar algorithms.