Search

Background: Quantifying and dealing with lack of consistency in administrative databases (namely, under-coding) requires tracking patients longitudinally without compromising anonymity, which is often a challenging task. Objective: This study aimed to (i) assess and compare different hierarchical clustering methods on the identification of individual patients in an administrative database that does not easily allow tracking of episodes from the same patient; (ii) quantify the frequency of potential under-coding; and (iii) identify factors associated with such phenomena. Method: We analysed the Portuguese National Hospital Morbidity Dataset, an administrative database registering all hospitalisations occurring in Mainland Portugal between 2011–2015. We applied different approaches of hierarchical clustering methods (either isolated or combined with partitional clustering methods), to identify potential individual patients based on demographic variables and comorbidities. Diagnoses codes were grouped into the Charlson an Elixhauser comorbidity defined groups. The algorithm displaying the best performance was used to quantify potential under-coding. A generalised mixed model (GML) of binomial regression was applied to assess factors associated with such potential under-coding. Results: We observed that the hierarchical cluster analysis (HCA) + k-means clustering method with comorbidities grouped according to the Charlson defined groups was the algorithm displaying the best performance (with a Rand Index of 0.99997). We identified potential under-coding in all Charlson comorbidity groups, ranging from 3.5% (overall diabetes) to 27.7% (asthma). Overall, being male, having medical admission, dying during hospitalisation or being admitted at more specific and complex hospitals were associated with increased odds of potential under-coding. Discussion: We assessed several approaches to identify individual patients in an administrative database and, subsequently, by applying HCA + k-means algorithm, we tracked coding inconsistency and potentially improved data quality. We reported consistent potential under-coding in all defined groups of comorbidities and potential factors associated with such lack of completeness. Conclusion: Our proposed methodological framework could both enhance data quality and act as a reference for other studies relying on databases with similar problems. [ABSTRACT FROM AUTHOR]

Clustering is an unsupervised machine learning method grouping data samples into clusters of similar objects, used as a system support tool in numerous applications such as banking customers profiling, document retrieval, image segmentation, and e-commerce recommendation engines. The effectiveness of several clustering techniques is sensible to the initialization parameters, and different solutions have been proposed in the literature to overcome this limitation. They require high computational memory consumption when dealing with big data. In this paper, we propose the application of a recent object detection Deep Learning model (YOLO-v5) for assisting the initialization of classical techniques and improving their effectiveness on two-variate datasets, leveraging the accuracy and reducing dramatically the memory and time consumption of classical clustering methods. [ABSTRACT FROM AUTHOR]

This survey rigorously explores contemporary clustering algorithms within the machine learning paradigm, focusing on five primary methodologies: centroid-based, hierarchical, density-based, distribution-based, and graph-based clustering. Through the lens of recent innovations such as deep embedded clustering and spectral clustering, we analyze the strengths, limitations, and the breadth of application domains—ranging from bioinformatics to social network analysis. Notably, the survey introduces novel contributions by integrating clustering techniques with dimensionality reduction and proposing advanced ensemble methods to enhance stability and accuracy across varied data structures. This work uniquely synthesizes the latest advancements and offers new perspectives on overcoming traditional challenges like scalability and noise sensitivity, thus providing a comprehensive roadmap for future research and practical applications in data-intensive environments. [ABSTRACT FROM AUTHOR]

As cloud computing has grown in popularity, several different models and deployment strategies have emerged, fulfilling specific needs of different users. Thus cloud service selection is a major task that combines business and technical aspects. Several cloud providers categorize their bundles to help users find the best fit. However, the policy behind this categorization is unclear, making it difficult to be consistent among providers. Therefore, this work explores the complication of selecting the optimal cloud service among numerous and comparable solutions and introduces a selection framework for cloud services based on clustering analysis, providing an overall size categorization of cloud services derived from the cloud market and making the categorization policy explicit and homogeneous across all providers. Finally, a real-word case study is presented, highlighting the application of the proposed framework. [ABSTRACT FROM AUTHOR]

The modern civil aircraft engine is a type of highly complex engineering system in design, manufacturing, and life-cycle management. They are constantly operated under extreme and critical conditions, and yet, high reliability and safety are top priorities in the civil aviation industry. To ensure top performance and efficiency in operations, engines follow a modular design. This article intends to apply the data-driven cluster analysis to real-life operation data for aircraft engine fleets, which provides a module maintenance priority recommendation solution to increase the efficiency of operations and best use of the engine values. [ABSTRACT FROM AUTHOR]

Pythagorean fuzzy set (PFS) is a broadening of intuitionistic fuzzy set that can represent the situations where the sum of membership and the non-membership values exceeds one. Adding parameterization to PFS, we obtain a structure named as Pythagorean fuzzy soft set (PFSS). It has a higher capacity to deal with vagueness as it captures both the structures of a PFS and a soft set. Several practical situations demand the measure of similarity between two structures, whose sum of membership value and non-membership value exceeds one. There are no existing tools to measure the similarity between PFSS and this paper put forward similarity measures for PFSS. An axiomatic definition for similarity measure is proposed for PFSS and certain expressions for similarity measure are introduced. Further, some theorems which express the properties of similarity measures are proved. A comparative study between proposed expressions for similarity measure is carried out. Also, a clustering algorithm based on PFSS is introduced by utilizing the proposed similarity measure. [ABSTRACT FROM AUTHOR]

The generation of collective preference assessments occupies a critical position in deriving accurate and reliable alternative rankings in the context of large-scale group decision making (LSGDM). In general, the collective opinion generation framework entails the following three phases, which are clustering analysis, weighting clusters, and preference aggregation. However, the clustering of experts has been frequently based on preference similarities among them without taking into account individual opinions in which knowledge elicitation plays a crucial role. The traditional collective opinion generation framework suffering from this drawback may result in unreliable decision outcomes. To this end, we propose an expertise-structure and risk-appetite-integrated two-tiered collective opinion generation framework to address this concern. The first tier of the two-tiered collective opinion generation framework divides the entire expert group into several subgroups based on individual expertise structures, which are extracted from hesitant fuzzy linguistic term set (HFLTS)-based preference assessments, and it then weighs the resulting clusters in accordance with the overall expertise levels. The second-tier clusters the first-tier subgroups conditioned on the indicator of individual assessment similarities and gathers the generated subgroup preference constructs in the use of the risk appetite-oriented power average operator. In addition, the notion of proportional HFLTSs was introduced to manifest collective evaluations in second-tier subgroups to eliminate information loss and distortion. The effectiveness and flexibility of the proposed collective opinion generation algorithm are eventually illustrated by a case study and a comparison analysis. [ABSTRACT FROM AUTHOR]

Comprehensive benchmarking of clustering algorithms is rendered difficult by two key factors: 1) the elusiveness of a unique mathematical definition of this unsupervised learning approach and 2) dependencies between the generating models or clustering criteria adopted by some clustering algorithms and indices for internal cluster validation. Consequently, there is no consensus regarding the best practice for rigorous benchmarking, and whether this is possible at all outside the context of a given application. Here, we argue that synthetic datasets must continue to play an important role in the evaluation of clustering algorithms, but that this necessitates constructing benchmarks that appropriately cover the diverse set of properties that impact clustering algorithm performance. Through our framework, HAWKS, we demonstrate the important role evolutionary algorithms play to support flexible generation of such benchmarks, allowing simple modification and extension. We illustrate two possible uses of our framework: 1) the evolution of benchmark data consistent with a set of hand-derived properties and 2) the generation of datasets that tease out performance differences between a given pair of algorithms. Our work has implications for the design of clustering benchmarks that sufficiently challenge a broad range of algorithms, and for furthering insight into the strengths and weaknesses of specific approaches. [ABSTRACT FROM AUTHOR]

Copyright of Electric Power Automation Equipment / Dianli Zidonghua Shebei is the property of Electric Power Automation Equipment Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In the era of Big Data, cluster analysis of high-dimensional data sets often suffers from the Curse of dimensionality. To overcome this problem, the dimensionality reduction through feature selection becomes inevitable. Co-clustering or two-way clustering is considered to be a more sophisticated tool than conventional one-way clustering. Moreover, the advent of multi-view learning shows that the subjects of a data set can be interpreted in many ways. Interestingly, a minimal number of existing feature selection algorithms take advantage of the co-clustering method and are designed to consider multi-view data. Motivated by this, in the current article, we propose a feature (gene) selection method for high dimensional gene expression (GE) data through a multi-objective optimization based multi-view Co-Clustering algorithm (named MMCo-Clus). A popular evolutionary technique – Non-dominated Sorting Genetic Algorithm-II (NSGA-II) has been utilized as the proposed method's underlying optimization strategy. First, we construct two views of a chosen data set, utilizing knowledge from two different biological data sources. Next, we develop the MMCo-Clus algorithm considering the constructed views to identify a set of “good” co-clustering solutions. Finally, based on a concept of consensus operation on the co-clustering outcome, a small number of most relevant and non-redundant features are extracted from the original feature-space. The reduced dimension formed by new feature-space causes to decrease the computational burden and noise level of original data. For experimental analysis, we have chosen three benchmark GE data sets. Our feature selection method's effectiveness is evaluated through sample-classification accuracy, accompanied by the cluster profile plot/Eisen plot/t-SNE plot, and biological/statistical significance test. A thorough comparative analysis with existing feature selection algorithms using external and internal evaluation metrics supports our proposed method's potency. [ABSTRACT FROM AUTHOR]

Exposure to environmental noise is related to negative health effects. To prevent it, the city councils develop noise maps and action plans to identify, quantify, and decrease noise pollution. Smart cities are deploying wireless acoustic sensor networks that continuously gather the sound pressure level from many locations using acoustics nodes. These nodes provide very relevant updated information, both temporally and spatially, over the acoustic zones of the city. In this paper, the performance of several data clustering techniques is evaluated for discovering and analyzing different behavior patterns of the sound pressure level. A comparison of clustering techniques is carried out using noise data from two large cities, considering isolated and federated data. Experiments support that Hierarchical Agglomeration Clustering and K-means are the algorithms more appropriate to fit acoustics sound pressure level data. [ABSTRACT FROM AUTHOR]

Complex networks have been widely adopted to represent a variety of complicated systems. Given a complex network, it is of great significance to perform accurate clustering for better understanding its intrinsic organization. To this end, a fuzzy-based clustering algorithm, i.e., FCAN, has been developed. Though effective, FCAN suffers from the disadvantage of slow convergence, which in return constrains its efficiency. To address this issue, this article proposes a fast fuzzy clustering algorithm, namely, F $^2$ CAN, which incorporates a generalized momentum method into FCAN. Its fast convergence is rigorous justified in theory. Empirical studies on five datasets from real applications demonstrate that F $^2$ CAN achieves a better performance when compared with FCAN and several state-of-the-art clustering algorithms in terms of convergence rate and clustering accuracy simultaneously. Hence, F $^2$ CAN has potential for addressing the clustering analysis of large-scale complex networks emerging from industrial applications. [ABSTRACT FROM AUTHOR]

In large-scale group decision making (LSGDM), appropriate clustering analysis is important to consensus reaching since it can reduce the interactive complexity among individuals. According to the traditional clustering method, a conflict may arise between the consensus reaching levels and total adjustment costs within clusters when individuals have different unit adjustment cost, which reflects their willingness to make concessions. Since this conflict may aggravate the consensus complexity, we propose a new $K$ -means clustering method that considers both preferences and the preference adjustment cost. The preference adjustment cost is attached to preferences with a parameter that can be determined by balancing this conflict. Because of such conflict, the proposed clustering algorithm can improve the similarity of intracluster individuals on the preference adjustment cost by offsetting some acceptable consensus reaching levels within clusters. According to the proposed clustering algorithm, individuals who have both similar preferences and adjustment willingness are classified into the same clusters. In this way, the moderator can provide similar compensation strategies for intracluster individuals, which will decrease the adjustment complexity. A practical case study of team construction examines the application of the proposed algorithm, and the related comparative analysis shows that it is convenient for managers to persuade individuals to reach a consensus under the improved clustering results. [ABSTRACT FROM AUTHOR]

Social network large-scale decision-making (SN-LSDM) has attracted much attention in the field of decision science. Clustering and consensus building are two important processes for solving SN-LSDM problems. Traditional clustering methods are typically based on the opinion similarity. When extending decision-making to social network contexts, we argue that trust relationships among decision-makers (DMs) serve as a reliable resource for clustering. This study proposes a trust Cop-Kmeans clustering algorithm, which is a semisupervised clustering technique using prior knowledge of trust constraints generated by trust relationships. The consensus-reaching process (CRP) is an effective tool for reducing differences of opinion. In general, the costs and resources associated with the CRP are limited. Therefore, minimum-cost consensus (MCC) models have been developed and used widely in various group decision-making contexts. As an important resource influencing decision-making, trust provides a common-sense perception that the opinion of a DM with a high trust degree is considered to be widely recognized by others. We hold that the DM with high trust and low consensus can reduce the consensus cost by voluntarily losing some trust. Consequently, an improved MCC model considering voluntary trust loss is developed. Finally, a numerical example is presented to illustrate the feasibility of the proposed clustering algorithm and consensus model. A comparative analysis is implemented to explore the influence of the trust constraint on clustering and the influence of trust loss on the CRP. [ABSTRACT FROM AUTHOR]

Channel multipath components (MPCs) clustering and cluster characterization are the prerequisite for the development of cluster based channel models. This article investigates the MPCs clustering based on machine learning (ML) and analyzes the cluster characteristics in typical high-speed railway (HSR) scenarios. A variational Bayesian Gaussian mixture model (GMM)-based algorithm is introduced to achieve the space–time clustering of MPCs, which incorporates the statistical characteristics of MPCs and can automatically determine the optimum number of clusters. Moreover, a novel density-based validity index is proposed for evaluating the MPCs clustering performance. The proposed validity index improves the traditional index by considering the intracluster density, which can be calculated according to the Graham scanning method and Green’s formula. In addition to synthetic datasets, realistic MPCs datasets collected in an HSR obstructed viaduct scenario are used for the performance evaluation of the clustering algorithm and clustering validity index. Based on clustering results in the measured scenario, static cluster characteristics, including cluster number, intercluster and intracluster delay spreads (DSs) and angle spreads (ASs), and dynamic cluster characteristics such as cluster lifetime and birth and death property of clusters, are extracted and analyzed. These results will be useful for cluster-based channel modeling in future HSR mobile communication systems. [ABSTRACT FROM AUTHOR]

Machine learning techniques have been mostly applied in gene expression cancer data. Socio-demographic data available in cancer registries could be explored, to get further insight into relationships between cancer types and their contributing factors. Moreover, less attention has been paid to analyse the mixed demographic data (numeric and categorical) from cancer registries and its association to the cancer types. The aim of this study is to identify subgroups of patients, having similar demographics characteristics, from the population based cancer registry in Brunei Darussalam and examine the prevalent cancer types in these subgroups. Four clustering algorithms are explored in the cluster analysis of Brunei Darussalam Cancer Registry; Two-step, Partitional Around Medoid, Agglomerative Hierarchical and Model-based. Gower distance was used for measuring similarity for mixed data types. To evaluate the clusters found; cluster distribution and Silhouette index were used for cluster quality, Cohen's Kappa Index for cluster stability and Cramer's V Coefficient for clinical relevance of clusters. Six distinct demographic subgroups were consistently found by three algorithms while model-based clustering solution were not considered for deeper analysis as highly imbalanced clusters were produced. The subgroups found have good quality clusters, moderate association with cancer types and high stability. The top three prevalent cancers associated with these subgroups were consistently identified using the three algorithms. Upon comparing the subgroups' ages during diagnosis, we identify possible screening behaviours of specific subgroups, suggesting for early screening awareness programmes. This study demonstrates the use of cluster analysis in a cancer registry to identify demographic subgroups that could suggest potential areas to develop targeted and improved healthcare management strategies. [ABSTRACT FROM AUTHOR]

As a result of emerging new techniques for scientific way of collecting data, we are able to accumulate data in large scale pertaining to various fields. One such method of data mining is Cluster analysis. Of all clustering algorithms, density-based clustering is better in terms of clustering quality and the way the data are handled. Density based clustering is advantageous over other clustering algorithms in the following ways -- arbitrary shaped clusters are formed; number of clusters need not be known and noise is handled. However, there are two main points that are critical in density-based clustering. Firstly, it is not effective while handling datasets of varied density. Secondly, the selection of input parameters ε and MinPts play a critical role in the quality of clustering. This paper proposes a model -- Automatic Generation of Parameters in Density-Based Spatial Clustering (AGP-DBSCAN) that aims at improving the density-based clustering by generating different candidate parameters. With these candidates, we will be able to handle both uniform density and varied density datasets. The results of experiments also look promising for different clustering datasets. [ABSTRACT FROM AUTHOR]

This article proposes a new class of Real Elliptically Skewed (RESK) distributions and associated clustering algorithms that integrate robustness and skewness into a single unified cluster analysis framework. Non-symmetrically distributed and heavy-tailed data clusters have been reported in a variety of real-world applications. Robustness is essential because a few outlying observations can severely obscure the cluster structure. The RESK distributions are a generalization of the Real Elliptically Symmetric (RES) distributions. To estimate the cluster parameters and memberships, we derive an expectation maximization (EM) algorithm for arbitrary RESK distributions. Special attention is given to a new robust skew-Huber M-estimator, which is also the approximate maximum likelihood estimator (MLE) for the skew-Huber distribution, that belongs to the RESK class. Numerical experiments on simulated and real-world data confirm the usefulness of the proposed methods for skewed and heavy-tailed data sets. [ABSTRACT FROM AUTHOR]

Identifying key research themes is an effective way to chart knowledge structures in a field of research and, in turn, stimulate new ideas and innovation. Most thematic analyses of a research field are based on some form of network analysis, e.g., citations and cowords, and most of these networks are made up of cohesive, highly overlapping groups of nodes. Based on the suggestion that the “universal features” of networks are to be found in these overlapping communities, we argue that these same communities in a keyword network should reveal the key research themes in a field of study. With no traditional method with which to test our theory, we combined a cluster percolation algorithm with a Word2Vec model, and in a case study on information science, we were not only able to detect the overlapping communities in a keyword similarity network, but we also found a new perspective on the importance of overlapping communities as a way to identify a field's key research themes. [ABSTRACT FROM AUTHOR]

Semi-supervised clustering is one of important research topics in cluster analysis, which uses pre-given knowledge as constraints to improve the clustering performance. While clustering a data set, people often get prior constraints from different information sources, which may have different representations and contents, to guide clustering process. However, most of existing semi-supervised clustering algorithms are based on single-source constraints and rarely consider to integrate multi-source constraints to enhance the clustering quality. To solve the problem, we analyze the relations among different types of constraints and propose an uniform representation for them. Based it, we propose a new semi-supervised clustering algorithm to find out a clustering that has good cluster structure and high consensus of all the sources of constraints. In the algorithm, we construct an optimization objective model and its solution method to achieve the aim. This algorithm can integrate multi-source constraints well to reduce the effect of incorrect constraints from single sources and find out a high-quality clustering. By the experimental studies on several benchmark data sets, we illustrate the effectiveness of the proposed algorithm, compared to other semi-supervised clustering algorithms. [ABSTRACT FROM AUTHOR]

Cluster analysis is an important technique in data analysis. However, there is no encompassing theory on scatterplots to evaluate clustering. Human visual perception is regarded as a gold standard to evaluate clustering. The cluster analysis based on human visual perception requires the participation of many probands, to obtain diverse data, and hence is a challenge to do. We contribute an empirical and data-driven study on human perception for visual clustering of large scatterplot data. First, we systematically construct and label a large, publicly available scatterplot dataset. Second, we carry out a qualitative analysis based on the dataset and summarize the influence of visual factors on clustering perception. Third, we use the labeled datasets to train a deep neural network for modeling human visual clustering perception. Our experiments show that the data-driven model successfully models the human visual perception, and outperforms conventional clustering algorithms in synthetic and real datasets. [ABSTRACT FROM AUTHOR]

In batch spawning fish, secondary growth oocytes (SGO) are recruited and spawned in successive cohorts, and multiple cohorts co-occur in spawning-capable females. So far, histological features such as the prevalence of cortical alveoli or yolk granules are conservatively used to distinguish oocytes in different developmental stages which do not necessarily correspond to different cohorts. In this way, valuable information about spawning dynamics remains unseen and consequently misleading conclusions might be drawn, especially for species with high spawning rates and increased overlapping among oocyte cohorts. We introduce a new method for grouping oocytes into different cohorts based on the application of the K-means clustering algorithm on the characteristics of cytoplasmic structures, such as the varying size and intensity of cortical alveoli and yolk granules in oocytes of different development. The method allowed the grouping of oocytes without the need of using oocyte diameter, and thus, a crucial histological bias dealing with the cutting angle and the orientation of reference points (e.g. nucleus) has been overcome. Using sardine, Sardina pilchardus , as a case study, the separation of cohorts provided new insight into the ovarian dynamics, indentifying successive recruitment of up to five oocyte cohorts between SGO recruitment and spawning. These results verified previous histological indications of the number of cohorts in sardine. Altogether, this method represents an improved tool to study species with complex ovarian dynamics. • A new method is presented for grouping oocytes of multiple spawning fish into distinct cohorts. • K-means clustering analysis was applied to the size and intensity of cortical alveoli and yolk globules. • Sardine was used as a case study and was shown to successively recruit up to five secondary oocyte cohorts. • This method can prove important for species with continuous oocyte production and high spawning frequency. [ABSTRACT FROM AUTHOR]

Contribution: This article uses student semester grade point average (GPA) as a measure of student success to take into account the temporal effects in student success. The findings highlight the student performance based on their demographic status and use of university resources such as financial aid. College campuses should not only increase current resources but also raise awareness of current resources and make them more accessible (e.g., easier to apply or automatic applications). This is especially important for some demographics such as Hispanic first-generation students. Background: Higher education institutions are facing retention and graduation problems. One way to improve this is by understanding why students are not academically successful. Research Questions: In this study, demographic information and past academic records were analyzed to understand patterns of student success. Methodology: A cluster analysis was conducted to understand groups of students based on academic performance and demographic information. Examples of these factors are enrollment status, financial status, first-generation status, housing status, and transfer status. For the purpose of getting more accurate results, the students were separated into two different groups according to their admission status: 1) freshman and 2) transfer. Findings: The results indicate Hispanic, first-generation, low-income students are not likely to apply for financial aid although they are eligible. They have lower GPA and take fewer units per semester than other students. This can cause delayed graduation and accumulating more debt. [ABSTRACT FROM AUTHOR]

Cluster analysis and outlier detection are two continuously rising topics in data mining area, which in fact connect to each other deeply. Cluster structure is vulnerable to outliers; inversely, outliers are the points belonging to none of any clusters. Unfortunately, most existing studies do not notice the coupled relationship between these two tasks and handle them separately. In this article, we consider the joint cluster analysis and outlier detection problem, and propose the Clustering with Outlier Removal (COR) algorithm. Specifically, the original space is transformed into a binary space via generating basic partitions. We employ Holoentropy to measure the compactness of each cluster without involving several outlier candidates. To provide a neat and efficient solution, an auxiliary binary matrix is introduced so that COR completely and efficiently solves the challenging problem via a unified K-means— with theoretical supports. Extensive experimental results on numerous data sets in various domains demonstrate the effectiveness and efficiency of COR significantly over state-of-the-art methods in terms of cluster validity and outlier detection. Some key factors including the basic partition number and generation strategy in COR with an application on abnormal flight trajectory detection are further analyzed for practical use. [ABSTRACT FROM AUTHOR]

The major evolution of the semantic web has become exchanging data between applications in all domains of activities. Based on this vision, different applications in recent days, e.g. in the fields of community web portals, social networking, e-learning, multimedia retrieval, etc. have been designed. Due to growing number of web services, clustering of web resources becomes a valuable tool for semantic web mining. Clustering of internet objects like Internet web pages’ intimate new methods for grouping correlated content for better understanding and satisfies massive user query results in web pages’ search. Hence, web pages clustering algorithms should be able to handle massive irregular content and discover knowledge regardless of the web page complexity. These algorithms vary depending on the characteristics and data types. So, choosing the most appropriate algorithm is not an easy process as it should be accurate in terms of time and space complexity. Therefore, this paper rigorously surveys the most important algorithms of different types used for web page clustering. In addition, a comparative analysis of all such algorithms are provided in terms of several parameters. Finally, a brief discussion is provided on why web page clustering is important in emerging era of Semantic Web of Thing (SWoT) applications. [ABSTRACT FROM AUTHOR]

Linear discriminant analysis (LDA) is one of commonly used supervised subspace learning methods. However, LDA will be powerless faced with the no-label situation. In this paper, the unsupervised LDA (Un-LDA) is proposed and first formulated as a seamlessly unified objective optimization which guarantees convergence during the iteratively alternative solving process. The objective optimization is in both the ratio trace and the trace ratio forms, forming a complete framework of a new approach to jointly clustering and unsupervised subspace learning. The extension of LDA into Un-LDA enables to not only complete unsupervised subspace learning via the explicitly presented subspace projection matrix but also simultaneously finish clustering and even clustering out-of-sample data via the explicitly presented transformation matrix. To overcome the difficulty in solving the non-convex objective optimization, we mathematically prove that the Un-LDA optimization in both forms can be transformed into the simple K-means clustering optimization when the subspace is determined. The Un-LDA optimization is eventually completed by alternatively optimizing the clusters using K-means and the subspace using the supervised LDA methods and iterating this whole process until convergence or stopping criterion. The experiments demonstrate that our proposed Un-LDA algorithms are comparable or even much superior to the counterparts. [ABSTRACT FROM AUTHOR]

Large-scale group decision-making (LSGDM) solution is usually based on the clustering analysis process (CAP) and consensus reaching process (CRP). However, CAP and CRP can be contradictory since CAP is performed based on the differences between potentially small groups and CRP is conducted to improve the overall similarity of a large group. To balance CAP and CRP, a dynamic clustering analysis process (DCAP) based on consensus evolution networks is proposed. A clustering algorithm proposed based on community detection method can be used to handle the diverse network structures with dynamic consensus thresholds. The clustering validity based on the intracluster consensus levels in subgroups and the intercluster consensus level among subgroups is evaluated. Then, the DCAP after each feedback adjustment round in CRP is reanalyzed. In such a way, effective clustering can also be found after a satisfying consensus is reached. Finally, a case study shows the availability of this approach and comparative analyses are provided to highlight the advantages from both theoretical and numerical perspectives. [ABSTRACT FROM AUTHOR]

This article is a position paper about models and algorithms that are generally called “stream clustering.” Semantics and methods used in this field are often co-opted from static clustering, but they do not serve well for streaming data analysis. Most “state-of-the-art” methods, such as sequential k-means, Birch, CluStream, DenStream, etc., acknowledge that the data are seen but once in real streaming analysis (e.g., intrusion detection, voter fraud, etc.). Interpretation of their outputs generally overlooks the fact that when the data cannot be saved, batch clustering ideas, such as preclustering assessment, partitioning, and cluster validity are not relevant. But in the current literature, the data, or some subset of it, are often saved for hindsight evaluation (we call this fake stream clustering). Our position? Useful analysis of real streaming data is in its infancy. We do not argue that current approaches to streaming clustering are wrong: rather, we regard them as transitional methods which will eventually lead to a new and useful paradigm for this type of computation. We think that this class of models and algorithms are actually classifiers, but with a special added component, viz., continuously updated cluster footprints of the instream processing. We need to carefully define the objectives of streaming analysis, and then choose terminology and methods that suit this evolving paradigm. [ABSTRACT FROM AUTHOR]

Cluster analysis is a data mining technique that has been widely used to exploit useful information in a great amount of data. Because of their evaluation mechanism based on an intracluster distance (ICD) function, traditional single-objective clustering algorithms are not appropriate for not-well-separated data. Specifically, they may easily result in the drop of the optimal solution accuracy on their late stages of search when dealing with the latter. To overcome the problem, in this paper a novel index reflecting the similarity of data within a cluster is presented and called intracluster cohesion (ICC). However, if a multiobjective method is used to cluster with ICD and ICC as the specified objectives, its clustering accuracy may depend on one’s experience. Motivated by these, we propose an accelerated two-stage particle swarm optimization (ATPSO) in which ${K}$ -means is utilized to accelerate particles’ convergence during the population initialization. Its clustering process consists of two stages. First, the main objective of minimizing ICD is to execute preliminary clustering; second, ICC is optimized to promote the clustering accuracy. Extensive experiments with the help of 17 open-source clustering sets in various geometric distributions are conducted. The results show that ATPSO outperforms PSO, ${K}$ -means PSO (KPSO), chaotic PSO (CPSO), and accelerated CPSO in terms of accuracy, and its efficiency is approximate to that of KPSO. Its convergence trend indicates that the adoption of the proposed ICC contributes to the clustering accuracy. Remarkably, compared with the Pareto-based multiobjective PSO, ATPSO can detect clusters more accurately and quickly through the proposed two-stage search. [ABSTRACT FROM AUTHOR]

A complex network is a network with non-trivial topological structures. It contains not just topological information but also attribute information available in the rich content of nodes. Concerning the task of cluster analysis in a complex network, model-based algorithms are preferred over distance-based ones, as they avoid designing specific distance measures. However, their models are only applicable to complex networks where the attribute information is composed of attributes in binary form. To overcome this disadvantage, we introduce a three-layer node-attribute-value hierarchical structure to describe the attribute information in a flexible and interpretable manner. Then, a new Bayesian model is proposed to simulate the generative process of a complex network. In this model, the attribute information is generated by following the hierarchical structure while the links between pairwise nodes are generated by a stochastic blockmodel. To solve the corresponding inference problem, we develop a variational Bayesian algorithm called TARA, which allows us to identify functionally meaningful clusters through an iterative procedure. Our extensive experiment results show that TARA can be an effective algorithm for cluster analysis in a complex network. Moreover, the parallelized version of TARA makes it possible to perform efficiently at its tasks when applied to large complex networks. [ABSTRACT FROM AUTHOR]

This paper focuses on the unsupervised detection of the Higgs boson particle using the most informative features and variables which characterize the "Higgs machine learning challenge 2014" data set. This unsupervised detection goes in this paper analysis through 4 steps: (1) selection of the most informative features from the considered data; (2) definition of the number of clusters based on the elbow criterion. The experimental results showed that the optimal number of clusters that group the considered data in an unsupervised manner corresponds to 2 clusters; (3) proposition of a new approach for hybridization of both hard and fuzzy clustering tuned with Ant Lion Optimization (ALO); (4) comparison with some existing metaheuristic optimizations such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). By employing a multi-angle analysis based on the cluster validation indices, the confusion matrix, the efficiencies and purities rates, the average cost variation, the computational time and the Sammon mapping visualization, the results highlight the effectiveness of the improved Gustafson-Kessel algorithm optimized withALO(ALOGK)to validate the proposed approach. Even if the paper gives a complete clustering analysis, its novel contribution concerns only the Steps (1) and (3) considered above. The first contribution lies in the method used for Step (1) to select the most informative features and variables. We used the t-Statistic technique to rank them. Afterwards, a feature mapping is applied using Self-Organizing Map (SOM) to identify the level of correlation between them. Then, Particle Swarm Optimization (PSO), a metaheuristic optimization technique, is used to reduce the data set dimension. The second contribution of thiswork concern the third step, where each one of the clustering algorithms as K-means (KM), Global K-means (GlobalKM), Partitioning AroundMedoids (PAM), Fuzzy C-means (FCM), Gustafson-Kessel (GK) and Gath-Geva (GG) is optimized and tuned with ALO. [ABSTRACT FROM AUTHOR]

Popular clustering algorithms based on usual distance functions (e.g., the Euclidean distance) often suffer in high dimension, low sample size (HDLSS) situations, where concentration of pairwise distances and violation of neighborhood structure have adverse effects on their performance. In this article, we use a new data-driven dissimilarity measure, called MADD, which takes care of these problems. MADD uses the distance concentration phenomenon to its advantage, and as a result, clustering algorithms based on MADD usually perform well for high dimensional data. We establish it using theoretical as well as numerical studies. We also address the problem of estimating the number of clusters. This is a challenging problem in cluster analysis, and several algorithms are available for it. We show that many of these existing algorithms have superior performance in high dimensions when they are constructed using MADD. We also construct a new estimator based on a penalized version of the Dunn index and prove its consistency in the HDLSS asymptotic regime. Several simulated and real data sets are analyzed to demonstrate the usefulness of MADD for cluster analysis of high dimensional data. [ABSTRACT FROM AUTHOR]

One of the most significant problems in cluster analysis is to determine the number of clusters in unlabeled data, which is the input for most clustering algorithms. Some methods have been developed to address this problem. However, little attention has been paid on algorithms that are insensitive to the initialization of cluster centers and utilize variable weights to recover the number of clusters. To fill this gap, we extend the standard fuzzy $k$ k -means clustering algorithm. It can automatically determine the number of clusters by iteratively calculating the weights of all variables and the membership value of each object in all clusters. Two new steps are added to the fuzzy $k$ k -means clustering process. One of them is to introduce a penalty term to make the clustering process insensitive to the initial cluster centers. The other one is to utilize a formula for iterative updating of variable weights in each cluster based on the current partition of data. Experimental results on real-world and synthetic datasets have shown that the proposed algorithm effectively determined the correct number of clusters while initializing the different number of cluster centroids. We also tested the proposed algorithm on gene data to determine a subset of important genes. [ABSTRACT FROM AUTHOR]

The structural graph clustering algorithm SCAN is a fundamental technique for managing and analyzing graph data. However, its high runtime remains a computational bottleneck, which limits its applicability. In this paper, we propose a novel interactive approach for tackling this problem on multicore CPUs. Our algorithm, called anySCAN, iteratively processes vertices in blocks. The acquired results are merged into an underlying cluster structure consisting of the so-called super-nodes for building clusters. During its runtime, anySCAN can be suspended for examining intermediate results and resumed for finding better results at arbitrary time points, making it an anytime algorithm which is capable of handling very large graphs in an interactive way and under arbitrary time constraints. Moreover, its block processing scheme allows the design of a scalable parallel algorithm on shared memory architectures such as multicore CPUs for speeding up the algorithm further at each iteration. Consequently, anySCAN uniquely is a both interactive and work-efficient parallel algorithm. We further introduce danySCAN an efficient bulk update scheme for anySCAN on dynamic graphs in which the clusters are updated in bulks and in a parallel interactive scheme. Experiments are conducted on very large real graph datasets for demonstrating the performance of anySCAN. They show its ability to acquire very good approximate results early, leading to orders of magnitude speedup compared to SCAN and its variants. Moreover, it scales very well with the number of threads when dealing with both static and dynamic graphs. [ABSTRACT FROM AUTHOR]

We give convergence guarantees for estimating the coefficients of a symmetric mixture of two linear regressions by expectation maximization (EM). In particular, we show that the empirical EM iterates converge to the target parameter vector at the parametric rate, provided the algorithm is initialized in an unbounded cone. In particular, if the initial guess has a sufficiently large cosine angle with the target parameter vector, a sample-splitting version of the EM algorithm converges to the true coefficient vector with high probability. Interestingly, our analysis borrows from tools used in the problem of estimating the centers of a symmetric mixture of two Gaussians by EM. We also show that the population of EM operator for mixtures of two regressions is anti-contractive from the target parameter vector if the cosine angle between the input vector and the target parameter vector is too small, thereby establishing the necessity of our conic condition. Finally, we give empirical evidence supporting this theoretical observation, which suggests that the sample-based EM algorithm may not converge to the target vector when initial guesses are drawn accordingly. Our simulation study also suggests that the EM algorithm performs well even under model misspecification (i.e., when the covariate and error distributions violate the model assumptions). [ABSTRACT FROM AUTHOR]

As one of the most widely used clustering techniques, the fuzzy $k$ -means (FKM) assigns every data point to each cluster with a certain degree of membership. However, conventional FKM approach relies on the square data fitting term, which is sensitive to the outliers with ignoring the prior information. In this paper, we develop a novel and robust fuzzy $k$ -means clustering algorithm, namely, joint learning of fuzzy $k$ -means and nonnegative spectral clustering with side information. The proposed method combines fuzzy $k$ -means and nonnegative spectral clustering into a unified model, which can further exploit the prior knowledge of data pairs such that both the quality of affinity graph and the clustering performance can be improved. In addition, for the purpose of enhancing the robustness, the adaptive loss function is adopted in the objective function, since it smoothly interpolates between $\ell _{1}$ -norm and $\ell _{2}$ -norm. Finally, experimental results on benchmark datasets verify the effectiveness and the superiority of our clustering method. [ABSTRACT FROM AUTHOR]

Clustering of remote sensing imagery is a tough task due to the particular and complex structure of remote sensing images and the shortage of known information. In this paper, we propose a fully automatic spectral–spatial fuzzy clustering method using an adaptive multiobjective memetic algorithm (AMOMA) for multispectral remote sensing imagery. This approach is made up of two automatic layers: an automatic determination layer and an automatic clustering layer. The first layer seeks the optimal number of clusters through a self-adaptive differential evolution algorithm. The second layer then takes advantage of the AMOMA for spectral–spatial clustering using the optimal number of clusters. The knee point from the Pareto front is then selected through the angle-based method in every generation, and we then compare the knee points between generations to output the final optimal solution. The effectiveness of the proposed method is verified by the experimental results obtained with three remote sensing data sets. [ABSTRACT FROM AUTHOR]

This paper presents a method for continuous indoor-outdoor environment detection on mobile devices based solely on Wi-Fi fingerprints. Detection of indoor-outdoor switching is an important part of identifying a user's context, and it provides important information for upper layer context aware mobile applications such as recommender systems, navigation tools, etc. Moreover, future indoor positioning systems are likely to use Wi-Fi fingerprints, and therefore Wi-Fi receivers will be on most of the time. In contrast to existing research, we believe that these fingerprints should be leveraged, and they serve as the basis of the proposed method. Using various machine learning algorithms, we train a supervised classifier based on features extracted from the raw fingerprints, clusters, and cluster transition graph. The contribution of each of the features to the method is assessed. Our method assumes no prior knowledge of the environment, and a training set consisting of the data collected for just a few hours on a single device is sufficient in order to provide indoor-outdoor classification, even in an unknown location or when using new devices. We evaluate our method in an experiment involving 12 participants during their daily routine, with a total of 828 h' worth of data collected by the participants. We report a predictive performance of the AUC (area under the curve) of 0.94 using the gradient boosting machine ensemble learning method. We show that our method can be used for other context detection tasks such as learning and recognizing a given building or room. [ABSTRACT FROM AUTHOR]

Clustering large volumes of high-dimensional data is a challenging task. Many clustering algorithms have been developed to address either handling datasets with a very large sample size or with a very high number of dimensions, but they are often impractical when the data is large in both aspects. To simultaneously overcome both the ‘curse of dimensionality’ problem due to high dimensions and scalability problems due to large sample size, we propose a new fast clustering algorithm called FensiVAT. FensiVAT is a hybrid, ensemble-based clustering algorithm which uses fast data-space reduction and an intelligent sampling strategy. In addition to clustering, FensiVAT also provides visual evidence that is used to estimate the number of clusters (cluster tendency assessment) in the data. In our experiments, we compare FensiVAT with nine state-of-the-art approaches which are popular for large sample size or high-dimensional data clustering. Experimental results suggest that FensiVAT, which can cluster large volumes of high-dimensional datasets in a few seconds, is the fastest and most accurate method of the ones tested. [ABSTRACT FROM AUTHOR]

Domain adaptation has been a primal approach to addressing the issues by lack of labels in many data mining tasks. Although considerable efforts have been devoted to domain adaptation with promising results, most existing work learns a classifier on a source domain and then predicts the labels for target data, where only the instances near the boundary determine the hyperplane and the whole structure information is ignored. Moreover, little work has been done regarding to multi-source domain adaptation. To that end, we develop a novel unsupervised domain adaptation framework, which ensures the whole structure of source domains is preserved to guide the target structure learning in a semi-supervised clustering fashion. To our knowledge, this is the first time when the domain adaptation problem is re-formulated as a semi-supervised clustering problem with target labels as missing values. Furthermore, by introducing an augmented matrix, a non-trivial solution is designed, which can be exactly mapped into a K-means-like optimization problem with modified distance function and update rule for centroids in an efficient way. Extensive experiments on several widely-used databases show the substantial improvements of our proposed approach over the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Efforts to mitigate public health concerns are showing encouraging results over the time but disparities across the geographic regions still exist within countries. Inadequate researches on the regional disparities of health indicators based on representative and comparable data create challenges to develop evidence-based health policies, planning and future studies in developing countries like Bangladesh. This study examined the disparities among districts on various maternal and child health indicators in Bangladesh. Cluster analysis–an unsupervised learning technique was used based on nationally representative dataset originated from Multiple Indicator Cluster Survey (MICS), 2012–13. According to our results, Bangladesh is classified into two clusters based on different health indicators with substantial variations in districts per clusters for different sets of indicators suggesting regional variation across the indicators. There is a need to differentially focus on community-level interventions aimed at increasing maternal and child health care utilization and improving the socioeconomic position of mothers, especially in disadvantaged regions. The cluster analysis approach is unique in terms of the use of health care metrics in a multivariate setup to study regional similarity and dissimilarity in the context of Bangladesh. [ABSTRACT FROM AUTHOR]

We address the problem of reducing the size of Craig’s interpolants used in SAT-based model checking. Craig’s interpolants are AND-OR circuits, generated by post-processing refutation proofs of SAT solvers. Being highly redundant, their compaction is typically tackled by reducing the proof graph and/or by exploiting standard logic synthesis techniques. In this paper, we propose a set of ad-hoc logic synthesis functions that, revisiting known logic synthesis approaches, specifically address speed and scalability. Though general and not restricted to interpolants, these techniques target the main sources of redundancy in combinational circuits. This paper includes an experimental evaluation, showing the benefits of the proposed techniques, on a set of benchmark interpolants arising from hardware model checking problems. [ABSTRACT FROM AUTHOR]

Highlights • Networks and graphs are naturally embedded in Euclidean hyperspheres. • Communicability embedding of networks/graphs reveals clusters in networks. • Nonmetric multidimensional scaling allows visualization of networks in 3D communicability space. • Communicability clusters of papers in a citation network reveal levels of mathematization. • Communicability clusters in a gene-gene network reveal genes that co-participate in cancer and other diseases. Abstract A complex network is a condensed representation of the relational topological framework of a complex system. A main reason for the existence of such networks is the transmission of items through the entities of these complex systems. Here, we consider a communicability function that accounts for the routes through which items flow on networks. Such a function induces a natural embedding of a network in a Euclidean high-dimensional sphere. We use one of the geometric parameters of this embedding, namely the angle between the position vectors of the nodes in the hyperspheres, to extract structural information from networks. First we propose a simple method for visualizing networks by reducing the dimensionality of the communicability space to 3-dimensional spheres. Secondly, we use clustering analysis to cluster the nodes of the networks based on their similarities in terms of their capacity to successfully deliver information through the network. After testing these approaches in benchmark networks and compare them with the most used clustering methods in networks we analyze two real-world examples. In the first, consisting of a citation network, we discover citation groups that reflect the level of mathematics used in their publications. In the second, we discover groups of genes that coparticipate in human diseases, reporting a few genes that coparticipate in cancer and other diseases. Both examples emphasize the potential of the current methodology for the discovery of new patterns in relational data. [ABSTRACT FROM AUTHOR]

The aim of the current study was to analyse performance differences of football players 2-years prior and the year after signing a new contract (the following year) while taking playing position, nationality, player’s role, team ability, and age into account. The sample was comprised of 249 players (n = 109 defenders, n = 113 midfielders; and n = 27 forwards) from four of the major European Leagues (Bundesliga, English FA Premier League, Ligue 1, and La Liga) during the seasons 2008 to 2015. The dependent variables studied were: shooting accuracy, defense (the sum of defensive actions, tackles, blocks, and interceptions), yellow cards, red cards, passing accuracy, tackle success, and minutes played per match. Two-step cluster analysis allowed classifying the sample into three groups of defenders (national important, foreign important, and less important players) and four groups of midfielders and forwards (national important, foreign important, national less important, and foreign less important players). Magnitude Based Inference (MBI) was used to test the differences between player’s performances during the years of analyses. The main results (very likely and most likely effects) showed better performance in the year prior to signing a new contract than the previous year for foreign important defenders (decreased number of red cards), national important midfielders (increased number of minutes played), foreign important forwards (increased minutes played and defense), and national important forwards (increased minutes played). In addition, performance was lower the year after signing the contract compared to the previous one for less important defenders (decreasing defense), national less important midfielders (decreased minutes played), and foreign less important forwards (decreased defense). On the other hand, the players showed better performance in defense and more minutes played the year after signing the contract for less important defenders, national less important midfielders, and foreign less important forwards. These results may assist coaches to decide on when a new contract should be signed or the duration of the contract. [ABSTRACT FROM AUTHOR]

Many real-world systems can be studied in terms of pattern recognition tasks, so that proper use (and understanding) of machine learning methods in practical applications becomes essential. While many classification methods have been proposed, there is no consensus on which methods are more suitable for a given dataset. As a consequence, it is important to comprehensively compare methods in many possible scenarios. In this context, we performed a systematic comparison of 9 well-known clustering methods available in the R language assuming normally distributed data. In order to account for the many possible variations of data, we considered artificial datasets with several tunable properties (number of classes, separation between classes, etc). In addition, we also evaluated the sensitivity of the clustering methods with regard to their parameters configuration. The results revealed that, when considering the default configurations of the adopted methods, the spectral approach tended to present particularly good performance. We also found that the default configuration of the adopted implementations was not always accurate. In these cases, a simple approach based on random selection of parameters values proved to be a good alternative to improve the performance. All in all, the reported approach provides subsidies guiding the choice of clustering algorithms. [ABSTRACT FROM AUTHOR]

Radio channel modeling has been an important research topic, since the performance of any communication system depends on channel characteristics. So far, most existing clustering algorithms are conducted based on the multipath components (MPCs) extracted by using a high-resolution parameter estimation approach, e.g., SAGE or MUSIC, etc. However, most of the estimation approaches require prior information to extract MPCs. Moreover, the high-resolution estimation approaches usually result in relatively high complexity, and thus, the clusters can only be identified by using an offline approach after the measurements. Therefore, a power-angle-spectrum (PAS) based clustering and tracking algorithm (PASCT) is proposed in this paper. First, a PAS is extracted from measurement data by using a Bartlett beamformer. For each PAS, the potential targets are selected from the background and separated into clusters by using image processing approaches. The recognized clusters are characterized by the following three attributes: size, position, and shape feature, where an orientation histogram is developed to describe the shape feature of the clusters. Moreover, a cost minimizing tracking approach based on Kuhn–Munkres method is proposed to accurately identify the clusters in non-stationary channels. The proposed PASCT algorithm is validated based on both simulations and measurements. It is found that the dominating clusters in both line-of-sight and non-line-of-sight environments can be well recognized and tracked with the proposed algorithm. By using the proposed algorithm, the dynamic changes of the clusters in real-time channel measurements, e.g., number, birth–death process, and size of the clusters, can be well observed. Through the experiments, the proposed algorithm can achieve fairly good accuracy on the cluster identification with lower complexity compared to the conventional solution. [ABSTRACT FROM AUTHOR]

This paper introduces a new evidential clustering method based on the notion of “belief peaks” in the framework of belief functions. The basic idea is that all data objects in the neighborhood of each sample provide pieces of evidence that induce belief on the possibility of such sample to become a cluster center. A sample having higher belief than its neighbors and located far away from the other local maxima is then characterized as cluster center. Finally, a credal partition is created by minimizing an objective function with the fixed cluster centers. An adaptive distance metric is used to fit for unknown shapes of data structures. We show that the proposed evidential clustering procedure has very good performance with an ability to reveal the data structure in the form of a credal partition, from which hard, fuzzy, possibilistic, and rough partitions can be derived. Simulations on synthetic and real-world datasets validate our conclusions. [ABSTRACT FROM AUTHOR]

When spectral clustering analysis is applied, a similarity matrix of data plays a vital role in both clustering performance and stability of clustering results. In order to enhance the clustering performance and maintain the stability of the clustering results, a new method to jointly learn the similarity matrix and the clustering structure, called the joint learning method (FSCM) of spectral clustering structure and fuzzy similarity matrix of data, is proposed in this paper. In FSCM, the capability of a double-index fuzzy C-means clustering algorithm is used to determine an appropriate fuzzy similarity between any pair of data points. A fuzzy similarity matrix of data is also determined by adaptively assigning fuzzy neighbors of data points so the spectral clustering structure of data can be found and the clustering stability of FSCM can be assured. Experimental results on synthetic and real datasets demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Network modules are used for diverse purposes, ranging from delineation of biogeographical provinces to the study of biotic interactions. We assess spatial scaling effects on modular structure, using a multi-step process to compare fish co-occurrence networks at three nested scales. We first detect modules with simulated annealing and use spatial clustering tests (interspecific distances among species’ range centroids) to determine if modules consist of species with broadly overlapping ranges; strong spatial clustering may reflect environmental filtering, while absence of spatial clustering may reflect positive interspecific relationships (commensalism or mutualism). We then use non-hierarchical, multivariate cluster analysis as an alternative method to identify fish subgroups, we repeat spatial clustering tests for the multivariate clusters, then compare spatial clustering results among modules and clusters. Next, we compare species lists within modules and clusters, and estimate congruence as the proportion of species assigned to the same groups by the two methods. Finally, we use a well-documented nest associate complex (fishes that deposit eggs in the gravel nests of a common host) to assess whether strong within-group associations may, in fact, reflect positive interspecific relationships. At each scale, 2–4 network modules were detected but a consistent relationship between scale and the number of modules was not observed. Significant spatial clustering was detected at all scales for network modules and multivariate clusters but was less prevalent at smaller scales. Congruence between modules and clusters was always < 90% and generally decreased as the number of groups increased. At all scales, the complete nest associate complex was completely preserved within a single network module, but not within a single multivariate cluster. Collectively, our results suggest that network modules are promising tools for studying positive interactions and that smaller scales may be preferable in this research. [ABSTRACT FROM AUTHOR]