Cluster Quality Assessment Research Articles

Optimized Spectral Clustering Methods For Potentially Divergent Biological Sequences

Various recent researches in bioinformatics demonstrated that clustering is a very efficient technique for sequence analysis. Spectral clustering is particularly efficient for highly divergent sequences and GMMs (Gaussian Mixture Models) are often able to cluster overlapping groups if given an adequately designed embedding. The current study used spectral embedding and Mixture Models for clustering potentially divergent biological sequences. The research approach resulted in a pipeline consisting of the following four steps. The first step consists of aligning the biological sequences. The pairwise affinity of the sequences is computed in the second step. Then the Laplacian Eigenmap embedding of the data is performed in the third step. Finally, the last step consists of a GMM-based clustering. Improving the quality of the generated clustering and the performance of this approach is directly related to the enhancement of each one of these four steps. The main contribution is proposing four GMM-based algorithms for automatically selecting the optimal number of clusters and optimizing the clustering quality. A clustering quality assessment method, based on phylogenetic trees, is also proposed. Moreover, a performance study and analysis have been conducted while testing different clustering methods and GMM implementations. Experimental results demonstrated the superiority of using the BIC (Bayesian Information Criterion) for selecting the optimal GMM configuration. Significant processing speed improvements were also recorded for the implementation of the proposed algorithms.

From A-to-Z review of clustering validation indices

Data clustering involves identifying latent similarities within a dataset and organizing them into clusters or groups. The outcomes of various clustering algorithms differ as they are susceptible to the intrinsic characteristics of the original dataset, including noise and dimensionality. The effectiveness of such clustering procedures directly impacts the homogeneity of clusters, underscoring the significance of evaluating algorithmic outcomes. Consequently, the assessment of clustering quality presents a significant and complex endeavor. A pivotal aspect affecting clustering validation is the cluster validity metric, which aids in determining the optimal number of clusters. The main goal of this study is to comprehensively review and explain the mathematical operation of internal and external cluster validity indices, but not all, to categorize these indices and to brainstorm suggestions for future advancement of clustering validation research. In addition, we review and evaluate the performance of internal and external clustering validation indices on the most common clustering algorithms, such as the evolutionary clustering algorithm star (ECA*). Finally, we suggest a classification framework for examining the functionality of both internal and external clustering validation measures regarding their ideal values, user-friendliness, responsiveness to input data, and appropriateness across various fields. This classification aids researchers in selecting the appropriate clustering validation measure to suit their specific requirements.

Optimizing Customer Relationship Management through Business Intelligence for Sustainable Business Practices

Amidst the dynamic landscape of contemporary business, the integration of Business Intelligence (BI) with Customer Relationship Management (CRM) emerges as a crucial paradigm for fostering sustainable business practices. This research investigates the synergy between BI-driven CRM strategies and sustainable operations, addressing the imperative to optimize customer relationships for sustainable business growth. Leveraging models such as BGNBD, and Gamma Gamma, and employing K-means clustering techniques, this study seeks to decode the intricate relationship between these strategies. The BGNBD model facilitates predictions of Customer Lifetime Value (CLTV), while the Gamma Gamma model estimates the Expected Average Profit, enabling a comprehensive understanding of customer behavior. Utilizing K-means clustering aids in customer segmentation, offering insights for targeted strategies. Visualization analyses, including the Elbow Method and Silhouette Plot, guide optimal cluster determination and cluster quality assessment. Ultimately, this research underscores the potential of BI-infused CRM approaches not only to drive profitability and enhance customer relationships but also to champion sustainable business practices. The findings provide a robust framework for businesses to craft and implement BI-enhanced CRM strategies, steering them toward sustainable growth while fostering customer-centricity and profitability in modern business environments.

A density-based statistical analysis of graph clustering algorithm performance

Abstract We introduce graph clustering quality measures based on comparisons of global, intra- and inter-cluster densities, an accompanying statistical significance test and a step-by-step routine for clustering quality assessment. Our work is centred on the idea that well-clustered graphs will display a mean intra-cluster density that is higher than global density and mean inter-cluster density. We do not rely on any generative model for the null model graph. Our measures are shown to meet the axioms of a good clustering quality function. They have an intuitive graph-theoretic interpretation, a formal statistical interpretation and can be tested for significance. Empirical tests also show they are more responsive to graph structure, less likely to breakdown during numerical implementation and less sensitive to uncertainty in connectivity than the commonly used measures.

Математические методы и алгоритмы интеллектуального анализа данных при оценке инвестиционной привлекательности IT-проектов

The article is concerned with developing mathematical support and algorithms for solving the problem of economic diagnostics of enterprises. IT-companies and start-ups (IT projects) that have special characteristics during the growth period were selected as the object of research. Based on the system analysis of data domain there has been developed a system of quantitative and qualitative characteristics to identify the economic state of the IT companies and start-ups in the external and internal environment. Scales of indices of different nature have been determined. Methods to introduce order and equivalence relations for the found peer companies have been given in order to compare their proximity to the analyzed company. Metrics used for comparing the companies are considered taking into account the quantitative and qualitative characteristics. The possibilities of distributing innovative IT projects using fuzzy clustering algorithms are considered. The comparative analysis of two basic algorithms - Fuzzy Classifier Means algorithm and Gustafson - Kessel algorithm - has been given. The clustering procedure for each algorithm is shown, as well as the graphic results of their operation. There was done the clustering quality assessment using a distribution coefficient, entropy of classification, and Hie-Beni index. It has been inferred that using Gustafson - Kessel algorithm provides better results for solving the problem of splitting IT projects for their economic diagnostics

Combining clustering and classification ensembles: A novel pipeline to identify breast cancer profiles

Breast Cancer is one of the most common causes of cancer death in women, representing a very complex disease with varied molecular alterations. To assist breast cancer prognosis, the classification of patients into biological groups is of great significance for treatment strategies. Recent studies have used an ensemble of multiple clustering algorithms to elucidate the most characteristic biological groups of breast cancer. However, the combination of various clustering methods resulted in a number of patients remaining unclustered. Therefore, a framework still needs to be developed which can assign as many unclustered (i.e. biologically diverse) patients to one of the identified groups in order to improve classification. Therefore, in this paper we develop a novel classification framework which introduces a new ensemble classification stage after the ensemble clustering stage to target the unclustered patients. Thus, a step-by-step pipeline is introduced which couples ensemble clustering with ensemble classification for the identification of core groups, data distribution in them and improvement in final classification results by targeting the unclustered data. The proposed pipeline is employed on a novel real world breast cancer dataset and subsequently its robustness and stability are examined by testing it on standard datasets. The results show that by using the presented framework, an improved classification is obtained. Finally, the results have been verified using statistical tests, visualisation techniques, cluster quality assessment and interpretation from clinical experts.

NEW MARKET SEGMENTATION METHODS USING ENHANCED (RFM), CLV, MODIFIED REGRESSION AND CLUSTERING METHODS

Behavioral segmentation is considered as one of the most important concepts of modern marketing. Traditional customer segmentation models require months of analytical work, resulting in discrete consumers’ insights that are outdated to match the dynamic body of the consumers they are meant to represent. Personalization and consumer experience are make or break factors for the retail industry. This study looks towards maximizing Consumer Lifetime Value (LTV) to accommodates the dynamics in consumer shopping behavior for a medium size retailer. using (LTV) matricto investigate behavioral changes in the consumer shopping history gaining knowledge from behavioral and demographic variables stored in POS database converted into RFM dataset format. In addition, this study applies soft clustering Fuzzy C-Means (FCM) and hard clustering Expectation Maximization (EM) algorithms to classify individual consumers exhibit similar purchase history into specific groups. For measuring the algorithms accuracy, weuse cluster quality assessment (CQA). The CQA shows EM algorithm scales much better than Fuzzyy C-Means algorithm with its ability to assign good initial points in the smaller dataset.

A MULTI-OBJECTIVE OPTIMIZATION PROBLEMS OF CLUSTERING PROTOCOLS FOR WIRELESS SENSOR NETWORKS USING META-HEURISTIC TECHNIQUES

Energy efficiency and prolonging the lifetime is a central issue in wireless sensor network. Clustering methods using meta-heuristic techniques are widely used to address this issue. Besides clustering, meta-heuristic techniques have been also applied to address several other key issues pertaining to WSN such as node deployment, coverage, optimal sensor deployment, topologycontrol, scheduling etc. These problems are formulated as single/multi T objective optimization problems. The overall objectives of the optimization are to minimize energy consumption and maximize the lifetime of the network. The article presents a mathematical formulation of a generic multi-objective optimization problems related to clustering. A generalized modelof meta-heuristic based clustering is proposed for its solution, taking into account the features of the WSN. The feature of the model comprises functional multi-purpose clustering quality assessment that takes into account the characteristics of spatial distribution and compactness of sensors, parameters of individual sensors and network characteristics. The article also gives a brief overview of different approaches to clustering methods. A comparative analysis of the improved protocols for heterogeneous WSN has been carried out, which showed the advantages of the proposed solutions.

Sub-classifying patients with mild traumatic brain injury: A clustering approach based on baseline clinical characteristics and 90-day and 180-day outcomes.

BackgroundThe current classification of traumatic brain injury (TBI) into “mild”, “moderate”, or “severe” does not adequately account for the patient heterogeneity that still exists within each of these categories. The objective of this study was to identify “sub-groups” of mild TBI (mTBI) patients based on data available at the time of the initial post-TBI patient evaluation and to determine if the sub-grouping correlates with patient outcomes at 90 and 180 days post-TBI.MethodsData from patients in the TRACK-TBI Pilot dataset who had a Glasgow Coma Scale (GCS) score of 13 to 15 at arrival to the Emergency Department and a closed head injury were included. Considering 53 clinical variables that are typically available during the initial evaluation of the patient with mild TBI, sparse heirarchial clustering with cluster quality assessment was used to identify the optimal number of patient sub-groups. Patient sub-groups were then compared for ten outcomes measured at 90 or 180 days post-TBI.ResultsAmongst the 485 patients with mTBI, optimal clustering was based on the inclusion of 12 clinical variables that divided the patients into 5 mild TBI sub-groups. Clinical variables driving the sub-clustering included: gender, employment status, marital status, TBI due to falling, brain CT scan result, systolic blood pressure, diastolic blood pressure, administration of IV fluids in the Emergency Department, alcohol use, tobacco use, history of neurologic disease, and history of psychiatric disease. These 5 mild TBI sub-groups differed in their 90 day and 180 day outcomes within several domains including global outcomes, persistence of TBI-related symptoms, and neuropsychological impairment.ConclusionsSub-groups of patients with mTBI can be identified according to clinical variables that are relatively easy to obtain at the time of initial patient evaluation. A patient’s sub-group assignment is associated with multidimensional patient outcomes at 90 and 180 days. These findings support the notion that there are clinically meaningful subgroups of patients amongst those currently classified as having mTBI.

Simultaneous optimisation of clustering quality and approximation error for time series segmentation

Time series segmentation is aimed at representing a time series by using a set of segments. Some researchers perform segmentation by approximating each segment with a simple model (e.g. a linear interpolation), while others focus their efforts on obtaining homogeneous groups of segments, so that common patterns or behaviours can be detected. The main hypothesis of this paper is that both objectives are conflicting, so time series segmentation is proposed to be tackled from a multiobjective perspective, where both objectives are simultaneously considered, and the expert can choose the desired solution from a Pareto Front of different segmentations. A specific multiobjective evolutionary algorithm is designed for the purpose of deciding the cut points of the segments, integrating a clustering algorithm for fitness evaluation. The experimental validation of the methodology includes three synthetic time series and three time series from real-world problems. Nine clustering quality assessment metrics are experimentally compared to decide the most suitable one for the algorithm. The proposed algorithm shows good performance for both clustering quality and reconstruction error, improving the results of other mono-objective alternatives of the state-of-the-art and showing better results than a simple weighted linear combination of both corresponding fitness functions.

PWLM[formula omitted]-based automatic performance model estimation method for HDFS write and read operations

There is a growing need for the development of an automatic performance model estimation method for Hadoop Distributed File System (HDFS) write and read (W/R) operations in order to deal with constant software improvement and updates, parameter configuration changes, hardware heterogeneity, and their Quality of Service (QoS) evaluation. Extant research based on single linear system model has a limited ability to explain the performance variations due to changes in HDFS parameters such as block size. These variations reveal some typical characteristics of nonlinear systems and are an obstacle in achieving effective automatic performance estimation. In order to deal with this challenge, a piecewise-linear multi-model modeling (PWLM3)-based automatic performance model estimation method is proposed for HDFS W/R performance. In the proposed method, a standard model base is built to standardize the model representation of every submodel. Moreover, a cluster quality assessment strategy is applied to evaluate the optimal number of submodels, and a submodel selection strategy is implemented to construct performance model candidates and improve the computation efficiency of the proposed method. In addition, Levenberg–Marquardt (LM) and Universal Global Optimization (UGO) algorithms are adopted to estimate the values of switch points and identify undetermined parameters of performance model candidates. Then the performance model is selected among these candidates according to Root Mean Squared Error (RMSE) indicator. Experimental results demonstrate that the PWLM3-based performance model provides a good understanding and description of nonlinear characteristics of HDFS W/R performance and achieves better identification precision than a single linear system model-based one.

Mathematical methods for monitoring the quality of education

The paper considers approaches to monitoring the quality of teaching. A list of existing classification algorithms is presented. The authors introduce the term of dynamic cluster quality assessment.

Using K-means method and spectral clustering technique in an outfitter’s value analysis

This study applies K-means method and spectral clustering technique in the customer data analysis of an outfitter in Taipei City, Taiwan. The data set contains transaction records of 551 customers from April 2004 to March 2006. The differences between the two clustering techniques mentioned here are significant. K-means method is more capable of dealing with linear separable input, while spectral clustering technique might have the advantage in non-linear separable input. Thus, it would be of interest to know which clustering technique performs better in a real-world case of evaluating customer value when the type of input space is unknown. By using cluster quality assessment, this study found that spectral clustering technique performs better than K-means method. To summarize the analysis, this study also suggests marketing strategies for each cluster based on the results generated by spectral clustering technique.

A case study of applying data mining techniques in an outfitter’s customer value analysis

This study applies K-means method, fuzzy c-means clustering method and bagged clustering algorithm to the analysis of customer value for an outfitter in Taipei, Taiwan. These three techniques bear similar philosophy for data classification. Thus, it would be of interest to know which clustering technique performs best in a real world case of evaluating customer value. Using cluster quality assessment, this study concludes that bagged clustering algorithm outperforms the other two methods. To conclude the analyses, this study also suggests marketing strategies for each cluster based on the results generated by bagged clustering technique.