Evolve systems using incremental clustering approach

Abstract

Incremental learning, software engineering, data mining, incremental clustering etc. are the major research areas and have received great attention in recent years. The intelligent system building cannot be possible without efficient knowledge management and machine learning. Since traditional machine learning approaches do not consider systemic interactions, they fail in complex decision scenarios. Increasing competition and the availability of multitudes of information and diminishing traditional entry barriers have left next generation businesses with the requirement to compete on knowledge grounds. This motivated my research in incremental learning. The effective reuse of organization data, fast and pragmatic learning based on context and augmentation of knowledge are some of the major research outcomes. This work introduces a new paradigm for machine learning and propose new framework for incremental clustering. Learning in dynamic scenarios, exploration and smart decision making are a few other facets of this work. With the information explosion, numerous information systems and complex decision scenarios there is need of truly smart learning systems and effective knowledge management. This work definitely builds the platform for dynamic and incremental learning. The stability of method is proved by various ways: it can handle all possible inputs—further mathematically prove that it converges—also prove same thing with experiments.