Phytochemical characteristics and the effect of methanolic extract of Erythrina variegata in the control of Aedes, Anopheles and Culex mosquito

Abstract :

Partitioning a dataset into identical clusters is a vital operation applied in data mining. Currently, there is an enormous effort is necessary to partition a big data into clusters. Clustering algorithms have developed as an alternative powerful meta-learning tool to precisely analyze the huge volume of data created by modern applications. In particular, their main objective is to classify data into clusters such that objects are gathered in the same cluster when they are similar permitting to specific metrics. This paper proposes a novel hybrid algorithm for big data clustering. The proposed method is designed with the help of Feature Subset Selection (FSS) algorithm and Distance based K-Means (DKM) algorithm. FFS helps to reduce the dimension of the dataset, it removes the irrelevant data from the dataset. A distance based k-means algorithm is best suited for implementing the clustering operation over big data. The experimental results for the proposed method is taken and compared with the existing method. The proposed method results better results in terms of two evaluation criteria such as error rate and adjusted rand index (ARI). The proposed method results better performance than the existing method and also provides lesser execution time.