Data clustering is one of the most popular techniques in data mining. It is a process of partitioning an unlabeled dataset into groups, where each group contains objects which are similar to each other with respect to a certain similarity measure and different from those of other groups. Clustering high-dimensional data is the cluster analysis of data which have anywhere from a few dozen to many thousands of dimensions. Such high-dimensional data spaces are often encountered in areas such as medicine, bioinformatics, biology, recommendation systems and the clustering of text documents. Many algorithms for large data sets have been proposed in the literature using different techniques. However, conventional algorithms have some shortcomings such as the slowness of their convergence and their sensitivity to initialization values. Particle Swarm Optimization (PSO) is a population-based globalized search algorithm that uses the principles of the social behavior of swarms. PSO produces better results in complicated and multi-peak problems. This paper presents a literature survey on the PSO algorithm and its variants to clustering high-dimensional data. An attempt is made to provide a guide for the researchers who are working in the area of PSO and high-dimensional data clustering.

A review on particle swarm optimization algorithm and its variants to clustering high-dimensional data

Clustering analysis is applied generally to Pattern Recognition, Color Quantization and Image Classification. It can help the user to distinguish the structure of data and simplify the complexity of data from mass information. The user can understand the implied information behind extracting these data. In real case, the distribution of information can be any size and shape. A particle swarm optimization algorithm-based technique, called PSO-clustering, is proposed in this article. We adopt the particle swarm optimization to search the cluster center in the arbitrary data set automatically. PSO can search the best solution from the probability option of the Social-only model and Cognition-only model[1, 2, 3J. This method is quite simple and valid and it can avoid the minimum local value. Finally, the effectiveness of the PSO-clustering is demonstrated on four artificial data sets.

[IEEE IEEE International Conference on Networking, Sensing and Control, 2004 - Taipei, Taiwan (March 21-23, 2004)] IEEE International Conference on Networking, Sensing and Control, 2004 - Particle swarm optimization algorithm and its application to clustering analysis

We proposed an improved binary artificial bee colony algorithm (IDisABC).We examined the proposed algorithm on dynamic clustering.Data and image clustering benchmark problems are chosen for experiments.The obtained results are compared with K-means, FCM, GA, DisABC, DCPSO. One of the most well-known binary (discrete) versions of the artificial bee colony algorithm is the similarity measure based discrete artificial bee colony, which was first proposed to deal with the uncapacited facility location (UFLP) problem. The discrete artificial bee colony simply depends on measuring the similarity between the binary vectors through Jaccard coefficient. Although it is accepted as one of the simple, novel and efficient binary variant of the artificial bee colony, the applied mechanism for generating new solutions concerning to the information of similarity between the solutions only consider one similarity case i.e. it does not handle all similarity cases. To cover this issue, new solution generation mechanism of the discrete artificial bee colony is enhanced using all similarity cases through the genetically inspired components. Furthermore, the superiority of the proposed algorithm is demonstrated by comparing it with the basic discrete artificial bee colony, binary particle swarm optimization, genetic algorithm in dynamic (automatic) clustering, in which the number of clusters is determined automatically i.e. it does not need to be specified in contrast to the classical techniques. Not only evolutionary computation based algorithms, but also classical approaches such as fuzzy C-means and K-means are employed to put forward the effectiveness of the proposed approach in clustering. The obtained results indicate that the discrete artificial bee colony with the enhanced solution generator component is able to reach more valuable solutions than the other algorithms in dynamic clustering, which is strongly accepted as one of the most difficult NP-hard problem by researchers.

Dynamic clustering with improved binary artificial bee colony algorithm

In recent years there has been an enormous increase in learning resources available online through massive open online courses and learning management systems. In this context, personalized resource recommendation has become an even more significant challenge, thereby increasing research in that direction. Recommender systems use ontology, artificial intelligence, among other techniques to provide personalized recommendations. Ontology is a way to model learners and learning resources, among others, which helps to retrieve details. This, in turn, generates more relevant materials to learners. Ontologies have benefits of reusability, reasoning ability, and supports inference mechanisms, which helps to provide enhanced recommendations. The comprehensive survey in this paper gives an overview of the research in progress using ontology to achieve personalization in recommender systems in the e-learning domain.

Review of ontology-based recommender systems in e-learning

Data clustering is one of important research topics of data mining. In this paper, we propose a new clustering algorithm based on ant colony optimization, called Ant Colony Optimization for Clustering (ACOC). At the core of the algorithm we use both the accumulated pheromone and the heuristic information, the distances between data objects and cluster centers of ants, to guide artificial ants to group data objects into proper clusters. This allows the algorithm to perform the clustering process more effectively and efficiently. Due to the nature of stochastic and population-based search, the ACOC can overcome the drawbacks of traditional clustering methods that easily converge to local optima. Experimental results show that the ACOC can find relatively good solutions.

An ACO-Based Clustering Algorithm

Clustering is a popular data analysis and data mining technique. It is the unsupervised classification of patterns into groups. Many algorithms for large data sets have been proposed in the literature using different techniques. However, conventional algorithms have some shortcomings such as slowness of the convergence, sensitive to initial value and preset classed in large scale data set etc. and they still require much investigation to improve performance and efficiency. Over the last decade, clustering with ant-based and swarm-based algorithms are emerging as an alternative to more traditional clustering techniques. Many complex optimization problems still exist, and it is often very difficult to obtain the desired result with one of these algorithms alone. Thus, robust and flexible techniques of optimization are needed to generate good results for clustering data. Some algorithms that imitate certain natural principles, known as evolutionary algorithms have been used in a wide variety of real-world applications. Recently, much research has been proposed using hybrid evolutionary algorithms to solve the clustering problem. This paper provides a survey of hybrid evolutionary algorithms for cluster analysis.

A survey: hybrid evolutionary algorithms for cluster analysis

Ontologies are currently emerging as representation techniques for overlapping compatibility context domains. The continuing need for more effective information retrieval has lead to the creation of the notions of the semantic web and personalized information management. Subsequently, the need for effective ontology visualization for design, management and browsing has arisen. Several ontology visualization tools have come out to strengthen the users' cognitive support. The primary goal of this paper is to present a survey on recently implemented ontology visualization tools and their contributions in the enrichment of users' cognitive support. This work also presents the preliminary results of an evaluation of three visualization tools to determine the suitability of each method for end user applications where ontologies are used as browsing aids.

A study on development of cognitive support features in recent ontology visualization tools

Ubiquitous Learning Environment (ULE) has been becoming a mobile and sensor based technology equipped environment that suits the modern world education discipline requirements for the past few years. Ambient Intelligence (AmI) makes much smarter the ULE by the support of optimization and intelligent techniques. Various efforts have been so far made as contributions to develop such smart environments. Hence a study is required to know the various smartness levels incorporated in those contributions. In this regard, first, a review on infrastructural developments for ubiquitous learning is reported in this article. Next, this report addresses various open issues, problems & requirements, technology challenges and potential improvements within the scope of Ubiquitous Learning Environment and Ambient Intelligence Assisted Learning Environment (AmIALE). It then also recommends an Internet-of-Things (IoT) enabled Ubiquitous Learning Environment for offering still smarter levels like energy efficiency, connectivity, special needs, Multi-modal Human Computer Interaction, Self-Optimization and Self-Discovery in an effective manner. We strongly believe that this work helps the researchers to start thinking in a different dimension to find solutions for various other current issues.

Ubiquitous and Ambient Intelligence Assisted Learning Environment Infrastructures Development - a review

Ontologies have been becoming popular knowledge representation methods very particularly for the areas of semantic web and smart environments. This paper focuses on design of "User Behaviour Ontologies" in Ambient Intelligence (Ami) environment using ontology creation tool. Sensors are installed in the environment so that users' behaviour will be sensed and stored by the devices in that environment. The stored behaviour data will be refined and formed as a temporal sequence. Then the ontology will be constructed through mapping relations between behaviour and related temporal sequence. This work demonstrated the different ways of ontological representations. The advantage of this approach is that it makes the cognitive support easier to the teachers.

http://ieeexplore.ieee.org/iel7/7430173/7435618/07435736.pdf

Development of ontologies for modelling user behaviour in Ambient Intelligence environment

/pdf/a-development-framework-for-social-media-interaction-in-54w8bmya67.pdf

Sivakumar Ramakrishnan

Papers

A survey: hybrid evolutionary algorithms for cluster analysis

A study on development of cognitive support features in recent ontology visualization tools

Ubiquitous and Ambient Intelligence Assisted Learning Environment Infrastructures Development - a review

Development of ontologies for modelling user behaviour in Ambient Intelligence environment

A Development Framework for Social Media Interaction in Learning Management System