A smart experience-based knowledge analysis system (SEKAS)

This article addresses the issues associated with using ever-increasing amounts of information and knowledge more effectively, and taking advantage of knowledge generated through experience. A hybrid structure, the Smart Experience-based Knowledge Analysis System (SEKAS), is put forward in this paper to address issues of knowledge management and use. SEKAS combines a set of experience knowledge structures (SOEKS) with multiple techniques to provide a comprehensive knowledge management approach capturing, discovering, reusing and storing knowledge for the users. The SEKAS integrates a novel Decisional DNA (DDNA) knowledge structure with the traditional web crawler technologies. DDNA, as a knowledge representation platform, can help deal with noisy and incomplete data, with learning from experience, and with making precise decisions and predictions in vague and fuzzy environments. The paper outlines the investigation of the combination of DDNA and feature selection algorithms to guarantee the future performance for prediction. The proposed approaches are general and extensible in terms of both designing novel algorithms, and in the application to other domains. The SEKAS integrates the evolutionary algorithm, NSGA-II, using experience that is derived from a former decision event, to improve the evolutionary algorithm's ability to find optimal solutions rapidly and efficiently. The SEKAS application to solve a travelling salesman's problem shows that this new proposed hybrid model can find optimal or close to true, Pareto-optimal solutions in a fast and efficient way.