Hierarchical Clustering Technique Research Articles

Swarm Intelligence-based Hierarchical Clustering for Identification of ncRNA using Covariance Search Model

Covariance Model (CM) has been quite effective in finding potential members of existing families of non-coding Ribonucleic Acid (ncRNA) identification and has provided ex-cellent accuracy in genome sequence database. However, it has significant drawbacks with family-specific search. An existing Hierarchical Agglomerative Clustering (HAC) technique merged overlapping sequences which is known as combined CM (CCM). However, the structural information will be discarded, and the sequence features of each family will be significantly diluted as the number of original structures increases. Additionally, it can only find members of the existing families and is not useful in finding potential members of novel ncRNA families. Furthermore, it is also important to construct generic sequence models which can be used to recognize new potential members of novel ncRNA families and define unknown ncRNA sequence as the potential members for known families. To achieve these objectives, this study proposes to implement Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) to ensure the CCMs have the best quality for every level of dendrogram hierarchy. This study will also apply distance matrix as the criteria to measure the compatibility between two CMs. The proposed techniques will be using five gene families with fifty sequences from each family from Rfam database which will be divided into training and testing dataset to test CMs combination method. The proposed techniques will be compared to the existing HAC in terms of identification accuracy, sum of bit-scores, and processing time, where each of these performance measurements will be statistically validated.

Hybrid Cluster based Collaborative Filtering using Firefly and Agglomerative Hierarchical Clustering

Recommendation Systems finds the user preferences based on the purchase history of an individual using data mining and machine learning techniques. To reduce the time taken for computation Recommendation systems generally use a pre-processing technique which in turn helps to increase high low performance and over comes over-fitting of data. In this paper, we propose a hybrid collaborative filtering algorithm using firefly and agglomerative hierarchical clustering technique with priority queue and Principle Component Analysis (PCA). We applied our hybrid algorithm on movielens dataset and used Pearson Correlation to obtain Top N recommendations. Experimental results show that the our algorithm delivers accurate and reliable recommendations showing high performance when compared with existing algorithms.

Equity clusters through the lens of realized semicorrelations

We rely on newly-developed realized semicorrelations constructed from high-frequency returns together with hierarchical clustering and cross-validation techniques to identify groups of individual stocks that share common features. Implementing the new procedures based on intraday data for the S&P 100 constituents spanning 2019-2020, we uncover distinct changes in the “optimal” groupings of the stocks coincident with the onset of the COVID-19 pandemic. Many of the clusters estimated with data post-January 2020 evidence clear differences from conventional industry type classifications. They also differ from the clusters estimated with standard realized correlations, underscoring the advantages of “looking inside” the correlation matrix through the lens of the new realized semicorrelations.

Livestock Informatics Toolkit: A Case Study in Visually Characterizing Complex Behavioral Patterns across Multiple Sensor Platforms, Using Novel Unsupervised Machine Learning and Information Theoretic Approaches.

Large and densely sampled sensor datasets can contain a range of complex stochastic structures that are difficult to accommodate in conventional linear models. This can confound attempts to build a more complete picture of an animal’s behavior by aggregating information across multiple asynchronous sensor platforms. The Livestock Informatics Toolkit (LIT) has been developed in R to better facilitate knowledge discovery of complex behavioral patterns across Precision Livestock Farming (PLF) data streams using novel unsupervised machine learning and information theoretic approaches. The utility of this analytical pipeline is demonstrated using data from a 6-month feed trial conducted on a closed herd of 185 mix-parity organic dairy cows. Insights into the tradeoffs between behaviors in time budgets acquired from ear tag accelerometer records were improved by augmenting conventional hierarchical clustering techniques with a novel simulation-based approach designed to mimic the complex error structures of sensor data. These simulations were then repurposed to compress the information in this data stream into robust empirically-determined encodings using a novel pruning algorithm. Nonparametric and semiparametric tests using mutual and pointwise information subsequently revealed complex nonlinear associations between encodings of overall time budgets and the order that cows entered the parlor to be milked.

3SHACC: Three stages hybrid agglomerative constrained clustering

Traditionally within the unsupervised learning paradigm, hierarchical and partitional clustering techniques have been shown to produce better results when provided with partial information, leading to a renewed attention towards this topic. Constrained clustering is a semi-supervised learning problem that combines classic clustering techniques with background knowledge given in the form of a set of constraints. In this paper, we propose to incorporate constraints into the clustering process in three phases: the first phase is devoted to quantify constraint relevance and to learn a metric matrix according to such relevance, a second phase computing similarities between instances by means of the reconstruction coefficient and pairwise distances, and a third stage performing agglomerative hierarchical clustering with a reward-style stepped affinity function favoring merges satisfying the higher possible number of constraints. Experimental results, supported by Bayesian statistical testing, show a consistent improvement in favor of our proposal over previous approaches to the constrained clustering problem.

Compositional data analysis of regional geochemical data in the Lhasa area of Tibet, China

It is now widely recognized that geochemical survey data are compositional in nature, with each component having relative importance as part of a whole. Compositional data analysis (CoDA) based on log-ratio transformations can be used to deal with the ‘sum to one’ data constraint. This study explored the use of two data-driven CoDA approaches, centered log-ratio (clr)-biplot analysis and a compositional-balance approach, to investigate associations between elements for potential mineral exploration in the Lhasa area of Tibet, China. Results revealed that (1) the compositional-balance approach, based on a hierarchical cluster and sequential binary partition (SBP) technique, well reflects the range of rocks and metal deposits in the area; (2) the clr-biplot indicates associated relationships between elements, and a consistency is found between principal component (PC1 and PC2) and key compositional balances (Balance 1 and 4); and (3) comparison with traditional integrated geochemical mapping, a knowledge-driven method, proves the validity of the compositional-balance and clr-biplot methods. These results provided metallogenic and petrogenetic information and crucial evidence supporting further geological and geochemical exploration. The improved knowledge demonstrates the importance of using a CoDA approach for geochemical data before performing further statistical analysis.

A clustering solution for analyzing residential water consumption patterns

Water utility companies in urban areas face two major challenges: ensuring there is enough water for everyone during prolonged drought and maintaining adequate water pressure during the hours of peak demand. These issues can be overcome by applying data analytics and machine learning to the data gathered from digital water meters. For water conservation and demand management strategies to be effective, utility companies need to gain a better understanding of consumer behaviours, habits and routines. To accomplish this goal, we adapted a clustering approach to reveal residential water consumption patterns within metered data. In the experiment, we used two data sets (engineered features data set as well as the times of use and weighted probabilities of use data set) based on the data collected over 10 months from 306 households in Melbourne, Australia. For the engineered features data set, first, we identified the number of optimal clusters. We then performed extensive experiments to find the best clustering approach in terms of performance evaluation and clustering quality. We chose the hierarchical agglomerative clustering technique based on the nature of the data and the objective of the study. We observed that for the engineered features data set, k-means is the best performing clustering technique after considering performance metrics. For the other data set, we found that the number of clusters varies based on the type of water-consumption event, type of day (i.e., weekday or weekend), profiling interval and probability of use. In addition, we observed that insight into tap-water usage could be used to determine the population’s adaptation of hygiene practices in an unprecedented time, such as the COVID-19 pandemic. Finally, we recommend that future clustering studies also employ aligned socio-demographic data and other key features.

Discretization of Fractional Operators: Analysis by Means of Advanced Computational Techniques

This paper studies the discretization of fractional operators by means of advanced clustering methods. The Grünwald–Letnikov fractional operator is approximated by series generated by the Euler, Tustin and generalized mean. The series for different fractional orders form the objects to be assessed. For this purpose, the several distances associated with the hierarchical clustering and multidimensional scaling computational techniques are tested. The Arc-cosine distance and the 3-dim multidimensional scaling produce good results. The visualization of the graphical representations allows a better understanding of the properties embedded in each type of approximation of the fractional operators.

Investigating spatial heterogeneity within fracture networks using hierarchical clustering and graph distance metrics

Abstract. Rock fractures organize as networks, exhibiting natural variation in their spatial arrangements. Therefore, identifying, quantifying, and comparing variations in spatial arrangements within network geometries are of interest when explicit fracture representations or discrete fracture network models are chosen to capture the influence of fractures on bulk rock behaviour. Treating fracture networks as spatial graphs, we introduce a novel approach to quantify spatial variation. The method combines graph similarity measures with hierarchical clustering and is applied to investigate the spatial variation within large-scale 2-D fracture networks digitized from the well-known Lilstock limestone pavements, Bristol Channel, UK. We consider three large, fractured regions, comprising nearly 300 000 fractures spread over 14 200 m2 from the Lilstock pavements. Using a moving-window sampling approach, we first subsample the large networks into subgraphs. Four graph similarity measures – fingerprint distance, D-measure, Network Laplacian spectral descriptor (NetLSD), and portrait divergence – that encapsulate topological relationships and geometry of fracture networks are then used to compute pair-wise subgraph distances serving as input for the statistical hierarchical clustering technique. In the form of hierarchical dendrograms and derived spatial variation maps, the results indicate spatial autocorrelation with localized spatial clusters that gradually vary over distances of tens of metres with visually discernable and quantifiable boundaries. Fractures within the identified clusters exhibit differences in fracture orientations and topology. The comparison of graph similarity-derived clusters with fracture persistence measures indicates an intra-network spatial variation that is not immediately obvious from the ubiquitous fracture intensity and density maps. The proposed method provides a quantitative way to identify spatial variations in fracture networks, guiding stochastic and geostatistical approaches to fracture network modelling.

Clustering diffused-particle method for scattering from large ensembles of electromagnetically polarizable particles

The Foldy-Lax equation is generalized for a medium that consists of particles with both electric and magnetic responses. The result is used to compute fields scattered from ensembles of particles. The computational complexity is reduced by hierarchical clustering techniques to enable simulations with on the order of ${10}^{10}$ particles. With so many particles we are able to see the transition to bulk media behavior of the fields. For nonmagnetic materials, the observable index, permittivity, and permeability of the effective bulk medium are in good agreement with the Clausius-Mossotti relation. The fields simulated for particles with both electric and magnetic responses are in good agreement with new analytical results for a generalized effective medium theory.

A Review on Clustering Techniques: Creating Better User Experience for Online Roadshow

Online roadshow is a relatively new concept that has higher flexibility and scalability compared to the physical roadshow. This is because online roadshow is accessible through digital devices anywhere and anytime. In a physical roadshow, organizations can measure the effectiveness of the roadshow by interacting with the customers. However, organizations cannot monitor the effectiveness of the online roadshow by using the same method. A good user experience is important to increase the advertising effects on the online roadshow website. In web usage mining, clustering can discover user access patterns from the weblog. By applying a clustering technique, the online roadshow website can be further improved to provide a better user experience. This paper presents a review of clustering techniques used in web usage mining, namely the partition-based, hierarchical, density-based, and fuzzy clustering techniques. These clustering techniques are analyzed from three perspectives: their similarity measures, the evaluation metrics used to determine the optimality of the clusters, and the functional purpose of applying the techniques to improve the user experience of the website. By applying clustering techniques in different stages of the user activities in the online roadshow website, the advertising effectiveness of the website can be enhanced in terms of its affordance, flow, and interactivity.

Assessment of precipitation trends and its implications in the semi-arid region of Southern India

Rainfall is a complex meteorological phenomenon with high spatio-temporal variability. These variations in hydrologic time series can be better analyzed using homogeneity and trend tests. In this paper, three homogeneity tests were applied to the annual rainfall series of 1980–2013 from 30 raingauge stations over the study area. The rainfall stations with homogeneous time series were subjected to cluster analysis (Agglomerative Hierarchical Clustering technique) and thereafter traditional trend tests (Mann-Kendall test and Spearman Rank-order Correlation test) and one emerging trend test (called ‘Innovative Trend test’) were applied to individual clusters. Based on the homogeneity test results, 27 rainfall stations with homogeneous time series were classified under six clusters. The results of the trend analysis revealed that a distinct pattern of increasing (decreasing) rainfall trend exists in the northern (southern) part of the study area. Although the traditional tests detected insignificant trends, the Innovative Trend (IT) test indicated the trends to be significant in all the clusters. However, the magnitude of trends detected by the traditional trend tests and non-traditional trend tests is similar for all the clusters. It is concluded that the emerging IT test is highly sensitive to the changes in the rainfall time series. The findings of this study suggest that the orography, population density, industrialization and agricultural intensity lead to a precipitation gradient in the study area.

Exploring the trend of walkability measures by applying hierarchical clustering technique

IntroductionOver the last two decades, many walkability attributes and variables have been considered in the evaluation of walkability scores with various methods and approaches. This study investigates the recent trend of walkability measures in terms of the scale of analysis, the case study location, the type of measurements, the data collection tool, the analysis method, and the research field. MethodsIn a review of existing literature, this paper analyzes and compares the walkability studies over the past two decades. A total of 75 articles from 40 journals and conference proceedings were selected through a systematic and sequential procedure. In order to explore the research trend among the reviewed articles, we applied a hierarchical clustering technique. A meta-analysis was performed to systematically assess the results of previous walkability studies to derive conclusions about the future direction of research. ResultsThe result of this review reveals that recent walkability studies in North America have prevailed in a street-level microscopic analysis while the others in Asia and Europe have focused on both large and small-scale walkability analysis. Advanced images and instrument audit tools are being applied nowadays in addition to traditional physical audit tools for measuring walkability. In recent years, urban design qualities objectively measuring at a street level (e.g., comfort/convenience, aesthetics, sidewalk design, etc.) have been widely used in determining walkability in lieu of the traditionally used attributes (e.g., roadway level of service). ConclusionThis study presents a future direction for walkability studies along with the recent trend in walkability measures and data analysis. Walkability studies have attracted significant attention more recently than in previous years. The recent trend of automated and systematic walkability data collection will eventually enable detailed analysis at a large scale and contribute to advancing walkability studies.

Consumer values in the Brazilian market for ethical cosmetics

PurposeThe purpose of this paper is to identify the personal values of consumers of ethical cosmetics in Brazil, using the resulting personality types to segment consumers for the development of strategies by Brazilian retailers and international players.Design/methodology/approachThis study administered the psychometric Values and Lifestyles Scale (VALS) via an online survey with 302 consumers of ethical cosmetics. First, exploratory factorial analysis was employed to identify the behavior of VALS' factors within the sample. Second, cluster analysis was performed using hierarchical clustering technique in order to link demographic variables and VALS' types to form specific consumer profiles.FindingsThe study found eight VALS types in the sample respondents but with a different configuration than the original factors. They were named, respectively, “Avant-garde”, “Oriented to fashion”, “Artisans”, “Committed to morality and religion”, “Leader of a group”, “Theoretical”, “Ingenious” and “Conservative”. Results indicated a demographically homogeneous sample with personality profiles mostly placed among VALS' original factors “Experiencer”, “Thinker” and “Innovator”.Research limitations/implicationsResults may vary within other cultural contexts and different means of investigation suggesting future research opportunities.Practical implicationsEthical demands concerning health and environmental preservation are no longer neglectable. The study of consumers' personal values can contribute to formulate suitable retail strategies for expected demands of consumers in the ethical cosmetics segment.Originality/valueThese findings are expected to provide resources for decision-makers, academics, practitioners and marketers concerning several points of sensitivity in their relationship with ethical consumers.

Project Performance Analysis Using Hierarchical Clustering Method

This paper deals with the project management performance analysis in automotive industry and usage of the advanced analytics to optimize the performance of the engineering projects within the vehicle manufacturing facility. The aim is to identify and mitigate the common causes of project delay (slippage) in various engineering projects that are independent of the technical discipline where a particular project belongs to. We have used the hierarchical clustering technique here to identify the sub-groups of similar projects and further deeply analysed those ones with the highest average slippage. Keywords—Cluster analysis, Data, Delay, Distance, Hierarchical clustering, Metric, Project, Slippage, Strategy

Age and gender differences in physical heart failure symptom clusters

BackgroundGender differences exist in structure and function of the heart resulting in HF symptom variation. Previous HF symptom cluster research described symptom clusters that were linked to functional status, mortality, quality of life and rehospitalization. Age and gender differences between cluster groups were described in one study. ObjectivesIdentify physical HF symptom clusters and explore age and gender differences between clusters. MethodsSecondary analysis study of adults with HF. Cluster analysis was conducted using hierarchical agglomerative clustering techniques. A pictorial dendrogram output displays clusters. ResultsThree symptom clusters were identified in this sample of 133 older HF patients that differed by gender (p = 0.04), age (p = 0.00) and beta blocker use (p = 0.01). Symptom clusters were consistent with worsening HF, acute HF and chronic HF. ConclusionSymptom clusters differ by age and gender. Education should be directed at increasing patient awareness of their individual symptom clusters.

Latent label mining for group activity recognition in basketball videos

Motion information has been widely exploited for group activity recognition in sports video. However, in order to model and extract the various motion information between the adjacent frames, existing algorithms only use the coarse video-level labels as supervision cues. This may lead to the ambiguity of extracted features and the omission of changing rules of motion patterns that are also important sports video recognition. In this paper, a latent label mining strategy for group activity recognition in basketball videos is proposed. The authors' novel strategy allows them to obtain the latent labels set for marking different frames in an unsupervised way, and build the frame-level and video-level representations with two separate levels of supervision signal. Firstly, the latent labels of motion patterns are digged using the unsupervised hierarchical clustering technique. The generated latent labels are then taken as the frame-level supervision signal to train a deep CNN for the frame-level features extraction. Lastly, the frame-level features are fed into an LSTM network to build the spatio-temporal representation for group activity recognition. Experimental results on the public NCAA dataset demonstrate that the proposed algorithm achieves state-of-the-art performance.

Identifying the characteristics of China’s maritime trading partners on the basis of bilateral shipping connectivity: a cluster analysis

ABSTRACT China is ranked as the number one maritime connected country in the world. This study attempts to analyse the characteristics of its 155 maritime trading partners. Five components of maritime connectivity have been used for the analysis. A hierarchical clustering technique is applied to group the trading partners into distinct meaningful clusters. As China’s main maritime competitors within the East Asian region, the analysis is also applied to Singapore and Hong Kong. The clustering technique produces three distinct clusters for China and Singapore and two for Hong Kong. The results reveal the importance of direct maritime connectivity, with China having the highest number of trading partners with zero transhipment (including its top 15 trading partners), compared to Singapore and Hong Kong. The analysis is further extended by applying ordered logistic regression. The results show that the logistic performance index (LPI) and exports/imports have a positive and significant influence on clusters. Conclusions are drawn as to the potential for implementing strategies and policy interventions on the basis of the emergent clusters, particularly the need for Hong Kong to improve all five components of maritime connectivity in order to compete with China and Singapore.

EXAMINING COMPETENCE IN ACTION RESEARCH OF BASIC EDUCATION TEACHERS IN CEBU CITY, PHILIPPINES

Background and Purpose: This study examined the self-perceived competence in action research components of basic education teachers in Cebu City as one form of identifying their challenges in doing AR. The results aimed to provide baseline information for the planned professional development program.
 
 Methodology: It employed a sequential explanatory design (QUAN→qual). It is characterized by collecting and analyzing quantitative data in the primary phase and then by collecting and analyzing qualitative data. There were 166 teachers who participated in the online survey. These teachers previously underwent professional development programs in designing AR projects, and two-thirds did an AR. Using their responses, they were grouped through a hierarchical clustering technique to create distinct groups of teachers sharing the similarity of competence or needs in the AR components. The cluster analysis yields four groups. Eventually, eight teachers were interviewed regarding their responses, which means two teachers represented each cluster.
 
 Findings: For very few teachers in cluster one (n=4), they regard selecting AR topic, planning the project, analyzing and presenting data, and integrating ethics as areas of non-difficulty while integrating technology, reflecting on AR, and communicating results as areas of difficulty. For the majority of the teachers belonging in cluster 2 (n=76), cluster 3 (n=37), and cluster 4 (n=49), all AR components are regarded as difficult, indicating all are critical areas for professional development.
 Contributions: A conventional way of conducting needs assessment of teachers’ competence in AR as a basis for professional development program is done through calculating the mean and standard deviation per AR competence or skills of all teachers participating in a survey. However, this method disregards the individual professional needs of teachers as it presents the general level of competence in each skill set. The professional needs may vary from one teacher to another. Thus, this study presents a novel way of examining teachers’ needs in AR by using cluster analysis to homogenously group participants according to the similarity of their responses or professional needs. This gives key reference points on which AR skills need to be improved for teachers belonging to the same group when planning a teacher development program in AR.
 
 Keywords: Action research, competence, in-service teacher, professional development, teacher research.
 
 Cite as: Cortes, S. T., Pineda, H. A., & Geverola, I. J. R. (2021). Examining competence in action research of basic education teachers in Cebu city, Philippines. Journal of Nusantara Studies, 6(2), 202-230. http://dx.doi.org/10.24200/jonus.vol6iss2pp202-230

Are current road investments exacerbating spatial inequalities inside European peripheral regions?

ABSTRACT The European Union has made considerable investments in transport infrastructures to reduce development gaps within and across territories and achieve a stronger regional cohesion. However, these economic efforts have not shown the expected effects, especially in peripheral regions, mainly due to the varied nature of their inner areas. This study aims to shed light on the existence of different types of areas inside peripheral regions that respond differently to the construction of new high-capacity roads, conditioning the achievement of cohesion goals. These disparities were explored through a detailed intraregional analysis of the peripheral Spanish Northwest Area over a 25-year period, through the identification of spatial categories that group homogenous areas in attention to three criteria: socio-economic development, spatial dynamics of urbanization and accessibility improvements. The application of a hierarchical clustering technique to different time scenarios and their comparison showed the existence of dynamic, stable and regressive areas, in terms of performance. Our findings reveal that with accessibility improvements regressive areas decreased and dynamic ones increased over time, but this transformation did not translate into an improvement in socio-economic intraregional cohesion. These results highlight the importance of considering intraregional diversity when formulating and implementing policies aimed at strengthening territorial cohesion.