Research On Parallel Clustering Algorithm Of Feature Hiding Big Data In Heterogeneous Networks

Abstract

Under the background of big data, network data presents a complex and multi structural feature. In the computer network, clustering is the key feature of big data hiding in heterogeneous network. In a word, the data chain in the network is more compact, and the data link between structures is more evacuation. As an important network data application method and traditional data development tool, clustering algorithm has been widely used in academia and society. Based on this, a parallel clustering algorithm for feature hiding big data in heterogeneous networks is proposed. On the premise of establishing the fuzzy equivalent constraint Association of hidden data, the heterogeneous measurement of mixed data is calculated, and online clustering is realized by reconstructing data structure. The experimental results show that the parallel clustering algorithm designed in this paper can achieve good clustering results in data sets, and can measure the differences between data and classes more accurately and reasonably. The new algorithm overcomes the shortcomings of traditional clustering algorithm which classifies attributes according to the overall size of data set or the dispersion degree within the cluster. Compared with other data clustering algorithms, the algorithm proposed in this paper has higher practicability and higher clustering quality.