Paired indices for clustering evaluation correction for agreement by chance

In the present paper we focus on the performance of clustering algorithms using indices of paired agreement to measure the accordance between clusters and an a priori known structure. We specifically propose a method to correct all indices considered for agreement by chance - the adjusted indices are meant to provide a realistic measure of clustering performance. The proposed method enables the correction of virtually any index - overcoming previous limitations known in the literature - and provides very precise results. We use simulated datasets under diverse scenarios and discuss the pertinence of our proposal which is particularly relevant when poorly separated clusters are considered. Finally we compare the performance of EM and KMeans algorithms, within each of the simulated scenarios and generally conclude that EM generally yields best results.

Quality indices for (practical) clustering evaluation

Clustering quality or validation indices allow the evaluation of the quality of clustering in order to support the selection of a specific partition or clustering structure in its natural unsupervised environment, where the real solution is unknown or not available. In this paper, we investigate the use of quality indices mostly based on the concepts of clusters` compactness and separation, for the evaluation of clustering results (partitions in particular). This work intends to offer a general perspective regarding the appropriate use of quality indices for the purpose of clustering evaluation. After presenting some commonly used indices, as well as indices recently proposed in the literature, key issues regarding the practical use of quality indices are addressed. A general methodological approach is presented which considers the identification of appropriate indices thresholds. This general approach is compared with the simple use of quality indices for evaluating a clustering solution.

Comparing clustering solutions: the use of adjusted paired indices

In the present paper we compare clustering solutions using indices of paired agreement. We propose a new method - IADJUST - to correct indices of paired agreement, excluding agreement by chance. This new method overcomes previous limitations known in the literature as it permits the correction of any index. We illustrate its use in external clustering validation, to measure the accordance between clusters and an a priori known structure. The adjusted indices are intended to provide a realistic measure of clustering performance that excludes agreement by chance with ground truth. We use simulated data sets, under a range of scenarios - considering diverse numbers of clusters, clusters overlaps and balances - to discuss the pertinence and the precision of our proposal. Precision is established based on comparisons with the analytical approach for correction specific indices that can be corrected in this way are used for this purpose. The pertinence of the proposed correction is discussed when making a detailed comparison between the performance of two classical clustering approaches, namely Expectation-Maximization (EM) and K-Means (KM) algorithms. Eight indices of paired agreement are studied and new corrected indices are obtained.

Clustering-based approach for categorizing pregnant women in obstetrics and maternity care

When a pregnant woman is guided to a hospital for obstetrics purposes, many outcomes are possible, depending on her current conditions. An improved understanding of these conditions could provide a more direct medical approach by categorizing the different types of patients, enabling a faster response to risk situations, and therefore increasing the quality of services. In this case study, the characteristics of the patients admitted in the maternity care unit of Centro Hospitalar of Porto are acknowledged, allowing categorizing the patient women through clustering techniques. The main goal is to predict the patients’ route through the maternity care, adapting the services according to their conditions, providing the best clinical decisions and a cost-effective treatment to patients. The models developed presented very interesting results, being the best clustering evaluation index: 0.65. The evaluation of the clustering algorithms proved the viability of using clustering based data mining models to characterize pregnant patients, identifying which conditions can be used as an alert to prevent the occurrence of medical complications.

Metrics for Association Rule Clustering Assessment

Issues related to association mining have received attention, especially the ones aiming to discover and facilitate the search for interesting patterns. A promising approach, in this context, is the application of clustering in the pre-processing step. In this paper, eleven metrics are proposed to provide an assessment procedure in order to support the evaluation of this kind of approach. To propose the metrics, a subjective evaluation was done. The metrics are important since they provide criteria to: (a) analyze the methodologies, (b) identify their positive and negative aspects, (c) carry out comparisons among them and, therefore, (d) help the users to select the most suitable solution for their problems. Besides, the metrics do the users think about aspects related to the problems and provide a flexible way to solve them. Some experiments were done in order to present how the metrics can be used and their usefulness.

Exploring the integration of data mining and data visualization

Due to the rapid advances in computing and sensing technologies, enormous amounts of data are being generated everyday in various applications. The integration of data mining and data visualization has been widely used to analyze these massive and complex data sets to discover hidden patterns. For both data mining and visualization to be effective, it is important to include the visualization techniques in the mining process and to generate the discovered patterns for a more comprehensive visual view. In this dissertation, four related problems: dimensionality reduction for visualizing high dimensional datasets, visualization-based clustering evaluation, interactive document mining, and multiple clusterings exploration are studied to explore the integration of data mining and data visualization. In particular, we 1) propose an efficient feature selection method (reliefF + mRMR) for preprocessing high dimensional datasets; 2) present DClusterE to integrate cluster validation with user interaction and provide rich visualization tools for users to examine document clustering results from multiple perspectives; 3) design two interactive document summarization systems to involve users efforts and generate customized summaries from 2D sentence layouts; and 4) propose a new framework which organizes the different input clusterings into a hierarchical tree structure and allows for interactive exploration of multiple clustering solutions.

Pulmonary tuberculosis in Continental Portugal 2000-2010: temporal trends clustering as a tool for control evaluation

Clustering analysis is a useful tool to detect and monitor disease patterns and, consequently, to contribute for an effective population disease management. Portugal has the highest incidence of tuberculosis in the European Union (in 2012, 21.6 cases per 100.000 inhabitants), although it has been decreasing consistently. Two critical PTB (Pulmonary Tuberculosis) areas, metropolitan Oporto and metropolitan Lisbon regions, were previously identified through spatial and space-time clustering for PTB incidence rate and risk factors. Identifying clusters of temporal trends can further elucidate policy makers about municipalities showing a faster or a slower TB control improvement.

Metrics to support the evaluation of association rule clustering

Many topics related to association mining have received attention in the research community, especially the ones focused on the discovery of interesting knowledge. A promising approach, related to this topic, is the application of clustering in the pre-processing step to aid the user to find the relevant associative patterns of the domain. In this paper, we propose nine metrics to support the evaluation of this kind of approach. The metrics are important since they provide criteria to: (a) analyze the methodologies, (b) identify their positive and negative aspects, (c) carry out comparisons among them and, therefore, (d) help the users to select the most suitable solution for their problems. Some experiments were done in order to present how the metrics can be used and their usefulness. © 2013 Springer-Verlag GmbH.

A score system for quality evaluation of RNA sequence tags: an improvement for gene expression profiling

Background: High-throughput molecular approaches for gene expression profiling, such as Serial Analysis of Gene Expression (SAGE), Massively Parallel Signature Sequencing (MPSS) or Sequencing-by-Synthesis (SBS) represent powerful techniques that provide global transcription profiles of different cell types through sequencing of short fragments of transcripts, denominated sequence tags. These techniques have improved our understanding about the relationships between these expression profiles and cellular phenotypes. Despite this, more reliable datasets are still necessary. In this work, we present a web-based tool named S3T: Score System for Sequence Tags, to index sequenced tags in accordance with their reliability. This is made through a series of evaluations based on a defined rule set. S3T allows the identification/selection of tags, considered more reliable for further gene expression analysis. Results: This methodology was applied to a public SAGE dataset. In order to compare data before and after filtering, a hierarchical clustering analysis was performed in samples from the same type of tissue, in distinct biological conditions, using these two datasets. Our results provide evidences suggesting that it is possible to find more congruous clusters after using S3T scoring system. Conclusion: These results substantiate the proposed application to generate more reliable data. This is a significant contribution for determination of global gene expression profiles. The library analysis with S3T is freely available at http://gdm.fmrp.usp.br/s3t/.S3T source code and datasets can also be downloaded from the aforementioned website.

Typical load profiles in the smart grid context – a clustering methods comparison

The present research paper presents five different clustering methods to identify typical load profiles of medium voltage (MV) electricity consumers. These methods are intended to be used in a smart grid environment to extract useful knowledge about customer’s behaviour. The obtained knowledge can be used to support a decision tool, not only for utilities but also for consumers. Load profiles can be used by the utilities to identify the aspects that cause system load peaks and enable the development of specific contracts with their customers. The framework presented throughout the paper consists in several steps, namely the pre-processing data phase, clustering algorithms application and the evaluation of the quality of the partition, which is supported by cluster validity indices. The process ends with the analysis of the discovered knowledge. To validate the proposed framework, a case study with a real database of 208 MV consumers is used.

Clustering an interval data set : are the main partitions similar to a priori partition?

This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Evaluation of the new advanced 15-loci MIRU-VNTR genotyping tool in Mycobacterium tuberculosis molecular epidemiology studies

Background. During the last few years, PCR-based methods have been developed to simplify and reduce the time required for genotyping Mycobacterium tuberculosis (MTB) by standard approaches based on IS6110-Restriction Fragment Length Polymorphism (RFLP). Of these, MIRU-12-VNTR (Mycobacterial interspersed repetitive units- variable number of tandem repeats) (MIRU-12) has been considered a good alternative. Nevertheless, some limitations and discrepancies with RFLP, which are minimized if the technique is complemented with spoligotyping, have been found. Recently, a new version of MIRU-VNTR targeting 15 loci (MIRU-15) has been proposed to improve the MIRU-12 format. Results. We evaluated the new MIRU-15 tool in two different samples. First, we analyzed the same convenience sample that had been used to evaluate MIRU-12 in a previous study, and the new 15-loci version offered higher discriminatory power (Hunter-Gaston discriminatory index [HGDI]: 0.995 vs 0.978; 34.4% of clustered cases vs 57.5%) and better correlation (full or high correlation with RFLP for 82% of the clusters vs 47%). Second, we evaluated MIRU-15 on a population-based sample and, once again, good correlation with the RFLP clustering data was observed (for 83% of the RFLP clusters). To understand the meaning of the discrepancies still found between MIRU-15 and RFLP, we analyzed the epidemiological data for the clustered patients. In most cases, splitting of RFLP-clustered patients by MIRU-15 occurred for those without epidemiological links, and RFLP-clustered patients with epidemiological links were also clustered by MIRU-15, suggesting a good epidemiological background for clustering defined by MIRU-15. Conclusion. The data obtained by MIRU-15 suggest that the new design is very efficient at assigning clusters confirmed by epidemiological data. If we add this to the speed with which it provides results, MIRU-15 could be considered a suitable tool for real-time genotyping.

MADAP, a flexible clustering tool for the interpretation of one-dimensional genome annotation data.

A recurring task in the analysis of mass genome annotation data from high-throughput technologies is the identification of peaks or clusters in a noisy signal profile. Examples of such applications are the definition of promoters on the basis of transcription start site profiles, the mapping of transcription factor binding sites based on ChIP-chip data and the identification of quantitative trait loci (QTL) from whole genome SNP profiles. Input to such an analysis is a set of genome coordinates associated with counts or intensities. The output consists of a discrete number of peaks with respective volumes, extensions and center positions. We have developed for this purpose a flexible one-dimensional clustering tool, called MADAP, which we make available as a web server and as standalone program. A set of parameters enables the user to customize the procedure to a specific problem. The web server, which returns results in textual and graphical form, is useful for small to medium-scale applications, as well as for evaluation and parameter tuning in view of large-scale applications, requiring a local installation. The program written in C++ can be freely downloaded from ftp://ftp.epd.unil.ch/pub/software/unix/madap. The MADAP web server can be accessed at http://www.isrec.isb-sib.ch/madap/.

On comparing two sequences of numbers and its applications to clustering analysis

A conceptual problem that appears in different contexts of clustering analysis is that of measuring the degree of compatibility between two sequences of numbers. This problem is usually addressed by means of numerical indexes referred to as sequence correlation indexes. This paper elaborates on why some specific sequence correlation indexes may not be good choices depending on the application scenario in hand. A variant of the Product-Moment correlation coefficient and a weighted formulation for the Goodman-Kruskal and Kendall`s indexes are derived that may be more appropriate for some particular application scenarios. The proposed and existing indexes are analyzed from different perspectives, such as their sensitivity to the ranks and magnitudes of the sequences under evaluation, among other relevant aspects of the problem. The results help suggesting scenarios within the context of clustering analysis that are possibly more appropriate for the application of each index. (C) 2008 Elsevier Inc. All rights reserved.