Mapping ISO 9001:2015 Compliance in Ecuadorian SMEs: An HJ-Biplot and Cluster Analysis

Rolling element bearing fault detection in industrial environments based on a K-means clustering approach

As sensor measurements emerge in urban water systems, data-driven unsupervised machine learning algorithms have drawn tremendous interest in event detection and hydraulic water level and flow prediction recently. However, most of them are applied in water distribution systems and few studies consider using unsupervised cluster analysis to group the time-series hydraulic-hydrologic data in stormwater urban drainage systems. To improve the understanding of how cluster analysis contributes to flooding location detection, this study compared the performance of K-means clustering, agglomerative clustering, and spectral clustering in uncovering time-series water depth dissimilarity. In this work, the water depth datasets are simulated by an urban drainage model and then formatted for a clustering problem. Three standard performance evaluation metrics, namely the silhouette coefficient index, Calinski–Harabasz index, and Davies–Bouldin index are employed to assess the clustering performance in flooding detection under various storms. The results show that silhouette coefficient index and Davies–Bouldin index are more suitable for assessing the performance of K-means and agglomerative clustering, while the Calinski–Harabasz index only works for spectral clustering, indicating these clustering algorithms are metric-dependent flooding indicators. The results also reveal that the agglomerative clustering performs better in detecting short-duration events while K-means and spectral clustering behave better in detecting long-duration floods. The findings of these investigations can be employed in urban stormwater flood detection at the specific junction-level sites by using the occurrence of anomalous changes in water level of correlated clusters as flood early warning for the local neighborhoods.

The article proposes a method for estimation of “correct” number of clusters. The stability methodology is used and quantative stability level assessment is introduced. This problem is important and common in cluster analysis and do not have unique criterion for all dataset types. The suggested approach solves one more problem. Often in socio-economics information analysis there are situations when initial numerical data contains various kinds of inaccuracies (measurement errors, intentional misrepresentations, errors of calculation, errors of mathematical model and other possible sources of errors). In this regard, it is important to choose those classifications of study objects which have the stability property with respect to random noise. The work is aimed to get a reliable and accurate clustering which is stable with respect to random perturbations and solves “ill posed” problems in clustering analysis, i.e. to find suggested number of clusters. In current paper the probabilistic approach to problem resolving is being offered. The variability frequency based on random perturbation is introduced and examined as a main metric for assessing clustering results. This estimation can be used for different clustering algorithms and their stability indices can be compared without additional procedures. The experiment on artificial data using the k-mean clustering approach is carried out.

WB-index: A sum-of-squares based index for cluster validity

Location determination is one of the important factors considered in agroindustry development. This study aimed to overview the regional grouping of seaweed agroindustry development in Indonesia. The K-mean Cluster approach was carried out to classify 26 seaweed-producing areas based on several criteria, such as availability of raw materials, labour, markets, and infrastructure. Data processing was conducted using SPSS software and produced 4 significantly different clusters. This was confirmed by a Euclidean distance obtained from ANOVA results, with a significance level below 0.05. The four clusters obtained can be categorized as “Highly Developing”, “Developing”, “Develop Enough”, and “Less Developed. Among 26 seaweed producing areas, East Java Province was included in a Highly Developing cluster, 3 provinces (South Sulawesi, Central Sulawesi, and East Nusa Tenggara) were included in a Developing cluster, 18 provinces (Maluku, West Nusa Tenggara, North Kalimantan, Southeast Sulawesi, Sulawesi North Maluku, Central Java, West Sulawesi, East Kalimantan, Gorontalo, Bengkulu, North Sumatra, West Kalimantan, South Kalimantan, Papua, West Papua, Lampung and Bali) were categorized into a “Develop Enough” cluster and 4 provinces (Banten, DKI Jakarta, West Java, Riau Islands) were categorized into a “Less Developed”.

This study compares the K-Medoids and CLARA (Clustering Large Application) methods for livestock population data in Indonesian districts and cities. Calculating the distance between points and objects in the data, K-Medoids is a method for clustering based on data points (medoids). A larger dataset is divided into several samples for comparison in CLARA, an extension of the K-Medoids approach. The CLARA method analysis results show that three clusters are the ideal number. The ideal number of clusters in a K-Medoids cluster analysis is two. The Silhouette Score (SS), Davis-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI) are the metrics that are measured. The evaluation of the comparison results shows that the CLARA method has an SS value of 0.66, while K-Medoids has an SS value of 0.62. The comparison of the CLARA and K-Medoids approaches yielded DBI values of 1.38 and 1.92, respectively, and 197.54 and 132.73 for CHI. The findings indicate that, in comparison to the K-Medoids approach, the SS value for the CLARA method is closer to 1, and that the CHI value derived from the CLARA method is likewise greater. The K-Medoids approach has a higher DBI value than the CLARA method, where a lower DBI value denotes superior performance. The CLARA approach is the most effective way to do cluster analysis on livestock population data in Indonesian districts and cities, according to the findings.

This paper constructs an indicator system composed of inherent attributes and time characteristics of the line based on the line loss, and proposes a K-Means line loss cluster analysis model based on this indicator system. The line is classified according to the clustering results. The result is 314.51 on the CH index (Calinski Harabasz Index), 0.19 on the Silhouette Cofficient (Silhouette Cofficient), and a running time of 0.508s. Compared with the traditional algorithm, it is greatly improved. The field of line loss analysis has guiding significance.

Climate classification plays a fundamental role in understanding climatic patterns, particularly in the context of a changing climate. This study utilized hourly meteorological data from 36 major cities in China from 2011 to 2021, including 2 m temperature (T2), relative humidity (RH), and precipitation (PRE). Both original hourly sequences and daily value sequences were used as inputs, applying two non-hierarchical clustering methods (k-means and k-medoids) and four hierarchical clustering methods (ward, complete, average, and single) for clustering. The classification results were compared using two clustering evaluation indices: the silhouette coefficient and the Calinski–Harabasz index. Additionally, the clustering was compared with the Köppen–Geiger climate classification based on the maximum difference in intra-cluster variables. The results showed that the clustering method outperformed the Köppen–Geiger climate classification, with the k-medoids method achieving the best results. Our research also compared the effectiveness of climate classification using two variables (T2 and PRE) versus three variables, including the addition of hourly RH. Cluster evaluation confirmed that incorporating the original sequence of hourly T2, PRE, and RH yielded the best performance in climate classification. This suggests that considering more meteorological variables and using hourly observation data can significantly improve the accuracy and reliability of climate classification. In addition, by setting the class numbers to two, the clustering methods effectively identified climate boundaries between northern and southern China, aligning with China’s traditional geographical division along the Qinling–Huaihe River line.

One of the most important problems among the methodological issues discussed in cluster analysis is the identification of the correct number of clusters and the correct allocation of units to their natural clusters. The most widely used index to determine the optimal number of groups is the Calinski Harabasz index. As shown in this paper, the presence of atypical observations has a strong effect on this index and may lead to the determination of a wrong number of groups. Furthermore, in order to study the degree of belonging of each unit to each group it is standard practice to apply a fuzzy k-means algorithm. In this paper we tackle this problem using a robust and efficient approach based on a forward search algorithm. The method is applied on a data set containing performance evaluation indicators of Italian universities.

Using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model to obtain backward trajectories and then conduct clustering analysis is a common method to analyze potential sources and transmission paths of atmospheric particulate pollutants. Taking Qingdao (N36 E120) as an example, the global data assimilation system (GDAS 1°) of days from 2015 to 2018 provided by National Centers for Environmental Prediction (NCEP) is used to process the backward 72 h trajectory data of 3 arrival heights (10 m, 100 m, 500 m) through the HYSPLIT model with a data interval of 6 h (UTC 0:00, 6:00, 12:00, and 18:00 per day). Three common clustering methods of trajectory data, i.e., K-means, Hierarchical clustering (Hier), and Self-organizing maps (SOM), are used to conduct clustering analysis of trajectory data, and the results are compared with those of the HYSPLIT model released by National Oceanic and Atmospheric Administration (NOAA). Principal Component Analysis (PCA) is used to analyze the original trajectory data. The internal evaluation indexes of Davies–Bouldin Index (DBI), Silhouette Coefficient (SC), Calinski Harabasz Index (CH), and I index are used to quantitatively evaluate the three clustering algorithms. The results show that there is little information in the height data, and thus only two-dimensional plane data are used for clustering. From the results of clustering indexes, the clustering results of SOM and K-means are better than the Hier and HYSPLIT model. In addition, it is found that DBI and I index can help to select the number of clusters, of which DBI is preferred for cluster analysis.

The k-means algorithm is a partitional clustering method. K-Means has several advantages, including being easy to implement, having a high level of convergence and producing denser clusters. Meanwhile, the drawback is that it is difficult to determine the optimal number of clusters. The K-Means method will be used to solve problems in areas prone to flood disasters in North Sumatra. This research aims to find the optimal number of clusters with the Calinski Harabasz Index and Krzanowski And Lai Index based on the Cluster Tightness Measure (CTM) value. There are eleven variables used in this research. Based on the research results, it was concluded that the CTM CH result of 0.376 was smaller than the CTM KL of 0.7843. So it can be said that determining the optimal number of clusters using CH with k = 6 is better than KL with k = 2.

Despite the well-known benefits of the Mediterranean diet (MD), low adherence is observed globally, highlighting the need to investigate the underlying causes of this trend. Large-scale, periodically repeated surveys, such as the European Health Interview Survey (EHIS), could be useful for the investigation of the factors that influence adherence to healthy dietary patterns. National EHIS-3 in Greece was designed to collect data on the consumption of all food groups, making it possible to determine adherence to MD. This study aimed to investigate the socioeconomic factors that influence MD adherence by conducting a secondary data analysis from 7706 participants aged 15-85 years who participated in the 2019 national EHIS. MD adherence was evaluated by MedDietScore. Bivariate comparisons between the score tertiles and socioeconomic groups were performed. Logistic regression analyses were used to estimate the association between MedDietScore tertiles and demographic and socioeconomic factors, using high adherence as the reference category. Finally, cluster analysis was applied to identify the most significant factors in relation to the classification of the participants in MD adherence groups. Educational level, followed by income status, emerged as the most significant factor associated with adherence to MD. Individuals who have attained only primary education had a 3.80 times higher likelihood of presenting low MD adherence instead of high MD adherence compared to persons with tertiary education. Individuals in the lower income group had 2.53 times higher odds of being in the low MD adherence instead of high MD adherence than individuals in the higher income group. These relationships remained statistically significant after adjusting for confounding factors. The group that most frequently adhered to high MD (53.4%) comprised individuals who had tertiary education, higher income, and were couples without children. Conversely, those with primary or secondary education who are single parents or live in one-person households and fall into the lower and medium income groups are most likely to have low MD adherence (44.8%). Public policies to promote MD adherence should consider socioeconomic factors. Integrating questions to evaluate adherence to dietary patterns into EHIS would allow for future intercountry and longitudinal comparisons.

Research problemTraditional approaches to talent development in Chinese higher education are increasingly misaligned with the evolving demands of an innovation-driven economy. Despite growing national interest, Innovation and Entrepreneurship (IAE) education in Chinese universities remains in its formative stage, with significant challenges in delivering personalized, data-informed instruction to diverse student populations.MethodologyThis study introduces a K-Means Clustering (KMC) approach to classify university students based on shared entrepreneurial traits and learning needs. Data were collected through structured surveys assessing entrepreneurial mindset, self-efficacy, and engagement in IAE activities. The KMC algorithm grouped students into three distinct clusters corresponding to undergraduate (UG), postgraduate (PG), and expert-level learners.ResultsThe clustering analysis revealed discrepancies in instructional support across student levels. Specifically, 33% of expert-level students, 41% of PG-level students, and 40% of UG-level students reported insufficient monitoring and guidance in their IAE learning experiences. These findings highlight the need for differentiated pedagogical strategies aligned with student readiness and capabilities.ImplicationsBy applying unsupervised machine learning, the study demonstrates how AI-driven analytics can enhance the personalization and effectiveness of IAE education. The results support the integration of clustering models into educational planning, allowing universities to optimize curriculum delivery and better prepare students for participation in innovation-oriented economies.

With the rapid growth of online property rental and sale platforms, the prevalence of fake real estate listings has become a significant concern. These deceptive listings waste time and effort for buyers and sellers and pose potential risks. Therefore, developing effective methods to distinguish genuine from fake listings is crucial. Accurately identifying fake real estate listings is a critical challenge, and clustering analysis can significantly improve this process. While clustering has been widely used to detect fraud in various fields, its application in the real estate domain has been somewhat limited, primarily focused on auctions and property appraisals. This study aims to fill this gap by using clustering to classify properties into fake and genuine listings based on datasets curated by industry experts. This study developed a K-means model to group properties into clusters, clearly distinguishing between fake and genuine listings. To assure the quality of the training data, data pre-processing procedures were performed on the raw dataset. Several techniques were used to determine the optimal value for each parameter of the K-means model. The clusters are determined using the Silhouette coefficient, the Calinski-Harabasz index, and the Davies-Bouldin index. It was found that the value of cluster 2 is the best and the Camberra technique is the best method when compared to overlapping similarity and Jaccard for distance. The clustering results are assessed using two machine learning algorithms: Random Forest and Decision Tree. The observational results have shown that the optimized K-means significantly improves the accuracy of the Random Forest classification model, boosting it by an impressive 96%. Furthermore, this research demonstrates that clustering helps create a balanced dataset containing fake and genuine clusters. This balanced dataset holds promise for future investigations, particularly for deep learning models that require balanced data to perform optimally. This study presents a practical and effective way to identify fake real estate listings by harnessing the power of clustering analysis, ultimately contributing to a more trustworthy and secure real estate market.

The development of digital health technologies during the COVID-19 outbreak sets the foundation for different research initiatives, including cluster analysis to better present and interpret the COVID-19 dataset. One of the cluster analysis algorithms often used is the K-means that various researchers applied in different applications. The application includes, but is not limited to, analyzing tourism attractions and restaurants, developing a machine learning model to track the virus's progression, and categorizing a country's cities or provinces based on its COVID-19 records. This paper intends to extend the cluster analysis of the prior research by enhancing a novel clustering algorithm known as K-means Firefly, which has improved the conventional K-means. The proponents achieved their objectives and were able to solve the limitation of the algorithm by improving the initialization of the algorithm's data and clusters, specifically by employing Principal Component Analysis for dimension reduction of data, the Calinski-Harabasz Index for automatically determining the number of clusters, and K-means++ for obtaining the initial clusters. The enhanced algorithm clustered the cities in the Philippines based on their COVID-19 datasets containing 32 healthcare-related features. As a result, the algorithm can handle real-world datasets with multiple features by reducing its dimensionality. It can automatically determine the optimal number of clusters and the initial location of the centroids. In terms of internal validity metrics, the enhanced algorithm also performed better than the previous implementation, with a percentage difference of 90.16% for the Silhouette Coefficient, 72.99% for the Davies-Bouldin Index, and 68.98 for the Calinski-Harabasz Index. The proposed algorithm may use for various applications, such as data dashboards and real-time tracking websites, due to its extended dynamic features in handling datasets and producing substantial clusters.