Cluster ensembles have recently emerged as a powerful alternative to standard cluster analysis, aggregating several input data clusterings to generate a single output clustering, with improved robustness and stability. From the early work, these techniques held great promise; however, most of them generate the final solution based on incomplete information of a cluster ensemble. The underlying ensemble-information matrix reflects only cluster-data point relations, while those among clusters are generally overlooked. This paper presents a new link-based approach to improve the conventional matrix. It achieves this using the similarity between clusters that are estimated from a link network model of the ensemble. In particular, three new link-based algorithms are proposed for the underlying similarity assessment. The final clustering result is generated from the refined matrix using two different consensus functions of feature-based and graph-based partitioning. This approach is the first to address and explicitly employ the relationship between input partitions, which has not been emphasized by recent studies of matrix refinement. The effectiveness of the link-based approach is empirically demonstrated over 10 data sets (synthetic and real) and three benchmark evaluation measures. The results suggest the new approach is able to efficiently extract information embedded in the input clusterings, and regularly illustrate higher clustering quality in comparison to several state-of-the-art techniques.
Motivation: It is far from trivial to select the most effective clustering method and its parameterization, for a particular set of gene expression data, because there are a very large number of possibilities. Although many researchers still prefer to use hierarchical clustering in one form or another, this is often sub-optimal. Cluster ensemble research solves this problem by automatically combining multiple data partitions from different clusterings to improve both the robustness and quality of the clustering result. However, many existing ensemble techniques use an association matrix to summarize sample-cluster co-occurrence statistics, and relations within an ensemble are encapsulated only at coarse level, while those existing among clusters are completely neglected. Discovering these missing associations may greatly extend the capability of the ensemble methodology for microarray data clustering. Results: The link-based cluster ensemble (LCE) method, presented here, implements these ideas and demonstrates outstanding performance. Experiment results on real gene expression and synthetic datasets indicate that LCE: (i) usually outperforms the existing cluster ensemble algorithms in individual tests and, overall, is clearly class-leading; (ii) generates excellent, robust performance across different types of data, especially with the presence of noise and imbalanced data clusters; (iii) provides a high-level data matrix that is applicable to many numerical clustering techniques; and (iv) is computationally efficient for large datasets and gene clustering. Availability: Online supplementary and implementation are available at: http://users.aber.ac.uk/nii07/bioinformatics2010 Contact:nii07@aber.ac.uk; natthakan@mfu.ac.th Supplementary information:Supplementary data are available at Bioinformatics online.
2 January 2017 in Human-centric Computing and Information Sciences
The implementation of data mining is widely considered as a powerful instrument for acquiring new knowledge from a pile of historical data, which is normally left unstudied. This data driven methodology has proven effective to improve the quality of decision-making in several domains such as business, medical and complex engineering problems. Recently, educational data mining (EDM) has obtained a great deal of attention among educational researchers and computer scientists. In general, publications in the field of EDM focus on understanding student types and targeted marketing, using both descriptive and predictive models to maximize student retention. Inspired by previous attempts, this paper aims to establish the clustering approach as a practical guideline to explore student categories and characteristics, with the working example on a real dataset to illustrate analytical procedures and results.
26 April 2010 in IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)
The intuition of data reliability has recently been incorporated into the main stream of research on ordered weighted averaging (OWA) operators. Instead of relying on human-guided variables, the aggregation behavior is determined in accordance with the underlying characteristics of the data being aggregated. Data-oriented operators such as the dependent OWA (DOWA) utilize centralized data structures to generate reliable weights, however. Despite their simplicity, the approach taken by these operators neglects entirely any local data structure that represents a strong agreement or consensus. To address this issue, the cluster-based OWA (Clus-DOWA) operator has been proposed. It employs a cluster-based reliability measure that is effective to differentiate the accountability of different input arguments. Yet, its actual application is constrained by the high computational requirement. This paper presents a more efficient nearest-neighbor-based reliability assessment for which an expensive clustering process is not required. The proposed measure can be perceived as a stress function, from which the OWA weights and associated decision-support explanations can be generated. To illustrate the potential of this measure, it is applied to both the problem of information aggregation for alias detection and the problem of unsupervised feature selection (in which unreliable features are excluded from an actual learning process). Experimental results demonstrate that these techniques usually outperform their conventional state-of-the-art counterparts.
17 February 2015 in International Journal of Machine Learning and Cybernetics