An evolutionary technique based on K-means algorithm for optimal clustering in RN
Information Sciences—Applications: An International Journal
Feature Weighting in k-Means Clustering
Machine Learning
On the performance of artificial bee colony (ABC) algorithm
Applied Soft Computing
Protein Interaction Networks: Computational Analysis
Protein Interaction Networks: Computational Analysis
Information Sciences: an International Journal
Kernel-induced fuzzy clustering of image pixels with an improved differential evolution algorithm
Information Sciences: an International Journal
Information Sciences: an International Journal
Improving functional modularity in protein-protein interactions graphs using hub-induced subgraphs
PKDD'06 Proceedings of the 10th European conference on Principle and Practice of Knowledge Discovery in Databases
Clustering PPI data based on Improved functional-flow model through Quantum-behaved PSO
International Journal of Data Mining and Bioinformatics
Hi-index | 0.07 |
Recently there are a large number of clustering methods applied to discover modules in protein-protein interaction (PPI) networks. However, due to the small-world and scale-free properties of PPI networks, most of them do not work well. This paper proposed a novel artificial bee colony (ABC) clustering model based on propagating mechanism. The ABC model based on propagating mechanism consisted of three different functions of bees which were named after queen, drone, and brood. The queen was regarded as a cluster center, and the drones stood for the sorted nodes according to the descending order of the aggregation coefficient of edge connecting these nodes with the queen node. The queen mated with the drones in order to cluster PPI data. In the end, the brood which is well-developed would be regarded as the new queen and went on a new mating-flight until all the nodes had been visited. This model could automatically obtain the cluster number during the clustering procedure, and the time complexity was greatly reduced. The simulation experiments on MIPS dataset showed that it performed well in terms of several criteria such as precision, recall and running time.