Improvements in Flock-Based Collaborative Clustering Algorithms

Abstract

Abstract.Inspiration from nature has driven many creative solutions to challenging real life problems. Many optimization methods, in particular clustering algorithms, have been inspired by such natural phenomena as neural systems and networks, natural evolution, the immune system, and lately swarms and colonies. In this paper, we make a brief survey of swarm intelligence clustering algorithms and focus on the flocks of agents-based clustering and data visualization algorithm, (FClust). A few limitations of FClust are then discussed with proposed improvements. We thus propose the FClust-annealing algorithm that decreases the number of iterations needed to converge and improves the quality of resulting clusters. We also propose a (Kmeans+FClust) hybrid algorithm which decreases the complexity of FClust from quadratic to linear, with further improvements in the cluster quality. Experiments on both artificial and real data illustrate the workings of FClust and the advantages of our proposed variants.

Swarm intelligence for clustering dynamic data sets for web usage mining and personalization.

“…In both algorithms, the ideal distance was computed using (28). Our experiments showed that since Web usage data followed a power low distribution [129], (24) resulted in a grossly over-estimated similarity threshold and led to no cluster formation, as mentioned in Section 2.5.2.…”

“…Our hybrid approach reduces the quadratic complexity of FClust to linear complexity, and performs similarly to FClust, but has the advantage of fewer iterations for clustering large, high-dimensional data such as web usage data. Hybrid algorithms were tested on several datasets including UCI machine learning data sets and Web server logs and our experiments confirmed their superiority, both in terms of quality of the final results and computational costs [128,129].…”

“…More recently, an approach based on flocks of agents, known as FClust, was used for data clustering [129,128,117,116]. The flock-based approach holds a great promise, and being the most recent of all the swarm-based clustering approaches, it has been the least studied.…”

“…Self-organizing maps (SOM) are also a popular way for data visualization and some of its applications include the visualization of web-based social networks [119,143]. Lately, swarm intelligence has also been used for data visualization [128,129,117].…”

“…In time, similar agents start moving together and closer, forming clusters [129,128,117]. Moreover, the distance between the agents depends on the similarity between the users that are mapped to those agents.…”