Fuzzy optimization multi-objective clustering Ensemble model for multi-source data analysis

Abstract

In modern data analysis, multi-source data appears more and more in real applications. Different data sources provide information about different data. Therefore, multi-source data linking is important to improve the processing performance. However, in practice multi-source data is often heterogeneous, un- certain, and large. This issue is considered a major challenge from multi-source data. Ensemble is a universal machine learning model in which learning techniques can work in parallel, with big data. Clustering ensemble has been shown to outperform any standard clustering algorithm in terms of ac- curacy and robustness. However, most of the traditional clustering ensemble approaches are based on single-objective function and single-source dataIn this paper, we pro- pose a new clustering ensemble method for multi-source data analysis. We call the fuzzy optimized multi-objective clustering ensemble method - FOMOCE. Firstly, a clustering ensemble mathematical model based on the structure of multi-objective clustering function, multi-source data, and dark knowledge is introduced. Then, rules for extracting dark knowledge from the input data, clustering algorithms, and base clustering are designed and applied. Finally, a clustering ensemble algorithm is proposed for multi-source data analysis. Experiments were performed on benchmark data sets. The experimental results demonstrate the superior performance of the FOMOCE method compared with the existing clustering ensemble methods and multi-source clustering methods.