A Spark-Based Artificial Bee Colony Algorithm for Large-Scale Data Clustering
Yanjie Wang, Quan Qian · 2018
Clustering is one of the most common data analysis methods which aims to partition data into a certain number of clusters, so that the data within the same cluster are similar and dissimilar from data in other clusters. Our research goal is to find more efficient clustering algorithms for large-scale data. Spark is the most popular distributed computing platform which provides a series of high-level API to make high-performance parallel applications. The Spark-based artificial bee algorithm proposed in this paper combines the robust artificial bee colony algorithm with the powerful Spark framework, which is very suitable for clustering large-scale data. To verify the effectiveness of this method, we adopt KDD CUP 99 data, an open competition dataset as the experimental data. The experimental results illustrate that our algorithm can get a good clustering quality and almost ideal speedup compared with the serial algorithms.