Obtaining Evidence Model of an Expert System Based on Machine Learning in Cloud Environment

Abstract

The main research of machine learning is through computer simulation or realization of human learning behavior to continuously acquire new knowledge or skills, for improving their own performance. Cloud computing environment provides massive and complex data. The variability and uncertain factors of data are the major disturbance to machine accuracy of knowledge acquisition. This paper proposes a comprehensive optimize evidence obtaining model based on multi-structure method to solve the evidence obtaining uncertainty problem in machine learning. The proposed model adopts the placement mapping method of decomposition to keep the valid evidence nodes rapidly and form them into a set of support evidence. Experimental results show that our approach has immense potential as it offers significant improvement in the aspects of remove the redundancy uncertain impact factor rapidly, and keep the valid certain evidence, guaranteeing the associated knowledge hided in the database can be found accurately.