Best Number Of Clusters Research Articles

Variability in recovery following microdiscectomy and postoperative physiotherapy for lumbar radiculopathy: A latent class trajectory analysis

ObjectivesThe clinical course of lumbar radiculopathy following microdiscectomy and post-operative physiotherapy varies substantially. No prior studies assessed this variability by deriving outcome trajectories. The primary aims of this study were to evaluate the variability in long-term recovery after lumbar microdiscectomy followed by post-operative physiotherapy and to identify outcome trajectories. The secondary aim was to assess whether demographic, clinical characteristics and patient-reported outcome measures routinely collected at baseline could predict poor outcome trajectories. MethodsWe conducted a prospective cohort study with a 24-month follow-up. We included 479 patients with clinical signs and symptoms of lumbar radiculopathy confirmed by Magnetic Resonance Imaging findings, who underwent microdiscectomy and post-operative physiotherapy. Outcomes were leg pain and back pain measured with Visual Analogue Scales, and disability measured with the Roland-Morris Disability Questionnaire. Descriptive statistics were performed to present the average and the individual clinical course. A latent class trajectory analysis was conducted to identify leg pain, back pain, and disability outcome trajectories. The best number of clusters was determined using the Bayesian Information Criterion, Akaike’s information criteria, entropy, and overall interpretability. Prediction models for poor outcome trajectories were assessed using multivariable logistic regression analyses. ResultsSeveral outcome trajectories were identified. Most patients were assigned to the ‘large improvement’ trajectory (leg pain: 79.3%; back pain: 70.2%; disability: 59.5% of patients). Smaller proportions of patients were assigned to the ‘moderate improvement’ trajectory (leg pain: 7.9%; back pain: 10.6%; disability: 20.7% of patients), the ‘minimal improvement’ trajectory (leg pain: 4.9%, back pain: 6.7%, disability: 16.3% of patients) and the ‘relapse’ trajectory (leg pain: 7.9%; back pain: 12.5%; disability: 3.5%). Approximately one-third of patients (32.6%) belonged to one or more than one poor outcome trajectory. Patients with previous treatment (prior back surgery, injection therapy, and medication use) and those who had higher baseline pain and disability scores were more likely to belong to the poor outcome trajectories in comparison to the large improvement trajectories in back pain, leg pain and disability, and the moderate improvement trajectory in disability. The explained variance (Nagelkerke R2) of the prediction models ranged from 0.06 to 0.13 and the discriminative ability (Area Under the Curve) from 0.66 to 0.73. ConclusionThe clinical course of lumbar radiculopathy varied following microdiscectomy and post-operative physiotherapy, and several outcome trajectories could be identified. Although most patients were allocated to favorable trajectories, one in three patients was assigned to one or more poor outcome trajectories following microdiscectomy and post-operative physiotherapy for lumbar radiculopathy. Routinely gathered data were unable to predict the poor outcome trajectories accurately. Prior to surgery, clinicians should discuss the high variability and the distinctive subgroups that are present in the clinical course with their patients.

Regional Tourism Development in Nusa Tenggara Barat: Maximizing Local Economic Development

The diversity of each region causes different potentials in each region. The potential of the village can map how rich the area is, the advantages of the area, and the population and welfare. Tourism is one of them; this sector is potential for the area because it can lift its economy if it is adequately managed. Good management is born from the policies/regulations of the local government. Nusa Tenggara Barat is a province with many tourist attractions. However, from an economic and socio-cultural perspective, Nusa Tenggara Barat has yet to be able to compete with other major provinces in Indonesia, such as the Special Region of Yogyakarta (DIY). The 2018 Village Potential Data by BPS can assist the government in compiling efforts for the village's progress. In the process of data processing, especially big data, in-depth exploration is needed to produce meaningful insight. Clustering is one of the exploration techniques that can map areas in Nusa Tenggara Barat based on the tourism potential in each village. K-Prototypes are used in cases with mixed variables (numeric and categorical). Determination of the best number of clusters is using the silhouette index. It produced 5 clusters with their respective diversity. There are five clusters in Nusa Tenggara Barat by the villages based on tourism aspects and factors that support tourism. Cluster 3 is an ideal cluster, meaning tourism development in that cluster is complete. Cluster 5 has considerable potential in tourism because the supporting factors are analytically good. There are villages dispersed across Sumbawa Barat, Sumbawa, Lombok Tengah, Lombok Barat, Dompu, and Bima that are part of cluster 1. In Sumbawa Barat and Lombok Tengah, cluster 1 predominates numbers. The settlements in cluster 2 are then more prevalent in Sumbawa and Bima. Furthermore, Sumbawa, Dompu, and Bima have the highest concentrations of cluster 4. Unlike clusters 3 and 5, special attention should be paid to clusters 1, 2, and 4 in tourism development. Implications of this research are the government could take toward each cluster to increase the GDP-oriented service product, namely tourism; whether it is an improvement or reconstruction, clustering analysis works its role in learning the data to make the policy more focused.

ANALISIS CLUSTER FAKTOR PENUNJANG PENDIDIKAN MENGGUNAKAN ALGORITMA K-MEANS (STUDI KASUS: KABUPATEN KARAWANG)

Education is one of the means to create and improve the quality of better human resources. This is expected to improve human welfare. Based on data from the Ministry of Education and Culture, there are 4 sub-districts in Karawang Regency that do not have state high schools. This can result in difficulties for students who have financial deficiencies which can ultimately lead to dropping out of school. Then apart from that distance can also be an obstacle. The purpose of this study is to apply the clustering method for the distribution of educational supporting factors in Karawang District so that later this research can improve the quality of education evenly in Karawang District, not only concentrated in certain areas by paying attention to educational supporting factors. The algorithm used in this research is K-Means. The elbow method used in determining the optimal cluster supported by the silhouette method resulted in the best number of clusters being 2 clusters. The results of the clustering evaluation resulted in Davies-Bouldin Index (DBI) value of 0.408 and Silhouette Coefficient value of 0.747 (strong structure). Cluster 1 consists of 7 sub-districts and Cluster 2 consists of 23 sub-districts. Based on the results of clustering analysis, Cluster 1 has the average number of attributes of schools, teachers, classes, laboratories, libraries at all levels of education (SD, SMP, SMA, SMK, and SLB) with state status and the population shows higher results if compared with the average number of each attribute in Cluster 2. So it can be concluded that Cluster 1 is a high category and Cluster 2 is a low category.

Assessing the androgenic and metabolic heterogeneity in polycystic ovary syndrome using cluster analysis.

Some but not all women with polycystic ovary syndrome (PCOS) develop the metabolic syndrome (MS). The objective of this study was to determine if a subset of women with PCOS had higher androgen levels predisposing them to MSand whether routinely measured hormonal parameters impacted the metabolic syndrome score (siMS). We included data from a discovery (PCOS clinic data) and a replication cohort (Hull PCOS Biobank) and utilized eight routinely measured hormonal parameters in our clinics (free androgen index [FAI], sex hormone-binding globulin, dehydroepiandrosterone sulphate (DHEAS), androstenedione, luteinizing hormone [LH], follicular stimulating hormone, anti-Müllerian hormone and 17 hydroxyprogesterone [17-OHP]) to perform a K-means clustering (an unsupervised machine learning algorithm). We used NbClust Package in R to determine the best number of clusters. We estimated the siMS in each cluster and used regression analysis to evaluate the effect of hormonal parameters on SiMS. The study consisted of 310 women with PCOS (discovery cohort: n = 199, replication cohort: n = 111). The cluster analysis identified two clusters in both the discovery and replication cohorts. The discovery cohort identified a larger cluster (n = 137) and a smaller cluster (n = 62), with 31% of the study participants. Similarly, the replication cohort identified a larger cluster (n = 74) and a smaller cluster (n = 37) with 33% of the study participants. The smaller cluster in the discovery cohort had significantly higher levels of LH (7.26 vs. 16.1 IU/L, p < .001), FAI (5.21 vs. 9.22, p < .001), androstenedione (3.93 vs. 7.56 nmol/L, p < .001) and 17-OHP (1.59 vs.3.12 nmol/L, p < .001). These findings were replicated in the replication cohort. The mean (±SD) siMS score was higher in the smaller cluster, 3.1 (±1.1) versus 2.8 (±0.8); however, this was not statistically significant (p = .20). In the regression analysis, higher FAI (β = .05, p = .003) and androstenedione (β = .03, p = .02) were independently associated with a higher risk of SiMS score, while higher DHEAS levels were associated with a lower siMS score (β = -.07, p = .03) CONCLUSION: We identified a subset of women in our PCOS cohort with significantly higher LH, FAI, and androstenedione levels. We show that higher levels of androstenedione and FAI are associated with a higher siMS, while higher DHEAS levels were associated with lower siMS.

Density-Based Unsupervised Learning Algorithm to Categorize College Students into Dropout Risk Levels

Compliance with the basic conditions of quality in higher education implies the design of strategies to reduce student dropout, and Information and Communication Technologies (ICT) in the educational field have allowed directing, reinforcing, and consolidating the process of professional academic training. We propose an academic and emotional tracking model that uses data mining and machine learning to group university students according to their level of dropout risk. We worked with 670 students from a Peruvian public university, applied 5 valid and reliable psychological assessment questionnaires to them using a chatbot-based system, and then classified them using 3 density-based unsupervised learning algorithms, DBSCAN, K-Means, and HDBSCAN. The results showed that HDBSCAN was the most robust option, obtaining better validity levels in two of the three internal indices evaluated, where the performance of the Silhouette index was 0.6823, the performance of the Davies–Bouldin index was 0.6563, and the performance of the Calinski–Harabasz index was 369.6459. The best number of clusters produced by the internal indices was five. For the validation of external indices, with answers from mental health professionals, we obtained a high level of precision in the F-measure: 90.9%, purity: 94.5%, V-measure: 86.9%, and ARI: 86.5%, and this indicates the robustness of the proposed model that allows us to categorize university students into five levels according to the risk of dropping out.

Application of Soft-Clustering Analysis Using Expectation Maximization Algorithms on Gaussian Mixture Model

Research on soft-clustering has not been explored much compared to hard-clustering. Soft-clustering algorithms are important in solving complex clustering problems. One of the soft-clustering methods is the Gaussian Mixture Model (GMM). GMM is a clustering method to classify data points into different clusters based on the Gaussian distribution. This study aims to determine the number of clusters formed by using the GMM method. The data used in this study is synthetic data on water quality indicators obtained from the Kaggle website. The stages of the GMM method are: imputing the Not Available (NA) value (if there is an NA value), checking the data distribution, conducting a normality test, and standardizing the data. The next step is to estimate the parameters with the Expectation Maximization (EM) algorithm. The best number of clusters is based on the biggest value of the Bayesian Information Creation (BIC). The results showed that the best number of clusters from synthetic data on water quality indicators was 3 clusters. Cluster 1 consisted of 1110 observations with low-quality category, cluster 2 consisted of 499 observations with medium quality category, and cluster 3 consisted of 1667 observations with high-quality category or acceptable. The results of this study recommend that the GMM method can be grouped correctly when the variables used are generally normally distributed. This method can be applied to real data, both in which the variables are normally distributed or which have a mixture of Gaussian and non-Gaussian.

COMBING: Clustering in Oncology for Mathematical and Biological Identification of Novel Gene Signatures.

Precision medicine is a paradigm shift in healthcare relying heavily on genomics data. However, the complexity of biological interactions, the large number of genes as well as the lack of comparisons on the analysis of data, remain a tremendous bottleneck regarding clinical adoption. In this paper, we introduce a novel, automatic and unsupervised framework to discover low-dimensional gene biomarkers. Our method is based on the LP-Stability algorithm, a high dimensional center-based unsupervised clustering algorithm. It offers modularity as concerns metric functions and scalability, while being able to automatically determine the best number of clusters. Our evaluation includes both mathematical and biological criteria to define a quantitative metric. The recovered signature is applied to a variety of biological tasks, including screening of biological pathways and functions, and characterization relevance on tumor types and subtypes. Quantitative comparisons among different distance metrics, commonly used clustering methods and a referential gene signature used in the literature, confirm state of the art performance of our approach. In particular, our signature, based on 27 genes, reports at least 30 times better mathematical significance (average Dunn's Index) and 25% better biological significance (average Enrichment in Protein-Protein Interaction) than those produced by other referential clustering methods. Finally, our signature reports promising results on distinguishing immune inflammatory and immune desert tumors, while reporting a high balanced accuracy of 92% on tumor types classification and averaged balanced accuracy of 68% on tumor subtypes classification, which represents, respectively 7% and 9% higher performance compared to the referential signature.

Application of hierarchical clustering on electricity demand of electric vehicles for GEP problems

Increasing fossil fuel consumption and consequently the effects of greenhouse gases (GHGs) on the environment and economy are a major concern for all nations and governments. Electric vehicles (EVs) with plug-in capabilities have the potential to ease such problems. However, the extracted power from the grid for charging the EVs' batteries will significantly impact daily power demand. To satisfy the increasing demand and ensure generation capacity adequacy, the generation expansion planning (GEP) problem is solved to determine the investment decisions for electricity generation sources. Even though there are no centralized utilities for generation planning in most markets, there is still a need to realistically solve the GEP problems and find the optimal investment decisions to tailor the incentives used by most governments to guide the market. There is also a need for a tool to analyze the effect of different charging power levels, charging policies, and penetration levels. The main goal of this paper is to provide a tool to determine realistic optimal investment plans and evaluate different cases. It is also very important to consider the stochastic nature of the electricity demand in GEP problems. We propose a scenario-based stochastic programming model to incorporate the variability in the electricity demand due to EV charging through a set of scenarios generated by Monte Carlo Simulation. The methodology starts with applying a simulation method to generate the electricity demand of EVs by considering all the possible factors affecting EVs' demand. Each iteration of this simulation represents a possible demand profile as a result of penetrating the EVs into the market. Using all these demand profiles in GEP is preferable, but it is not computationally efficient. Computational tractability is achieved by using the clustering technique to reduce the size of such scenarios. We propose clustering methods to select a representative set from the data sets generated by the simulation and integrate EVs into GEP problems by using the selected set. The GEP models are defined to represent EVs' demand explicitly and then solved to imply the benefit of the suggested methods. The results show that GEP models with a representative set produce more realistic solutions than the GEP models including only average EVs demand. To select representative sets, different clustering techniques and distance measurements are used and compared with respect to their performances. Two different methods are defined to choose the best number of clusters: the silhouette coefficient method and the

Penerapan k-Means Clustering Berdasarkan Analisis RFM Terhadap Segmentasi Pembeli untuk Meningkatkan Strategi CRM

An industry requires a good strategy in running its business. Saga Bako is a small industry that sells various types of tobacco and its equipment. However, Saga Bako has not yet implemented a Customer Relationship Management (CRM) strategy in its service to buyers. It is necessary to segment customers to find out less profitable buyers and buyers who provide large profits. The use of data mining also contributes when segmenting customers through the use of purchase data. The methodology applied in this research is CRISP-DM with purchase data at Saga Bako from January to March 2022. The k-means algorithm is applied in the formation of clusters based on the Recency, Frequency, Monetary (RFM) model, with the help of Weka 3.8.5 tools. The Elbow method is used to determine the best number of clusters (k). The results obtained are from 47 buyers with 663 transaction data divided into three clusters, 26 low potential buyers, ten medium potential buyers, and 11 high potential buyers.

Pendekatan Clustering untuk Menganalisis Efisiensi dan Kinerja Mahasiswa Berdasarkan Data Menerapkan Metode K-Means

The purpose of this study is to cluster the efficiency and performance of students. This is because the academic community is currently faced with several challenges in terms of analyzing and evaluating the progress of a student's academic achievement. In the real world, classifying student performance is a scientifically challenging task. Recently, several studies have applied cluster analysis to evaluate student outcomes and used statistical techniques to divide their scores in relation to student performance. This approach, however, is not efficient. In this study, we combined two techniques, namely k-mean and elbow clustering algorithm to evaluate student performance. Based on this combination, the performance results will be more accurate in analyzing and evaluating the progress of student performance, the application of the Elbow method according to this study gives the best number of clusters to 3, and when the K-Means method is applied, data is generated that the number of students is 73 students, from 4 repetitions. There are 3 clusters, namely the category of "Achievable", "Potential for Achievement", and "Less Achievement", with the results of the "Achievable" cluster as many as 34 students with a percentage of 47.22%, the cluster "Potential for Achievement" as many as 24 students with a percentage of 33.33 %, and the "Less Achievement" cluster as many as 15 students with a percentage of 19.45%.

DATA CLUSTERING ON TEACHING AND LEARNING EVALUATION APPLICATION OVER PANDEMIC ERA

SIEPEL is an evaluation application for teaching and learning process in University of Bengkulu. It is mandatory for every student to fill the questionnaire before they can see the marking value for each subject each semester. This survey was designed to meet the requirements and expectations of students as educational service for customers. This data is very important to improve the quality of teaching and learning process for further policy and decision maker. However, the analysis of the data remains an open question as the size and the distribution of the data is become some issues to process the analysis. Here, we showed the new approach to analysis the data using K-Means Clustering to see the better distribution and understanding over the evaluation data. This paper used elbow method to find the best number of clusters to be implemented on the algorithm approach which results in four clusters of satisfaction values (unsatisfied, less satisfied, satisfied, and very satisfied). The result of this analysis was published based on website system to show the visualization of analysis. Furthermore, this research showed that the average value of evaluation result for 4 semester was very satisfied 6.50%; satisfied 43.89%; less satisfied 44.26%; and not satisfied 5.36%. The value of vary satisfied students was dropped from 20.47% to 0.12% by 2 years and the value for less satisfied was increased from 27.64% to 66.32%. This term was happened because of the pandemic era and the change on the process of learning and teaching on University of Bengkulu.

Peningkatan Ketersediaan Darah Sesuai Segmentasi Umur Menggunakan K-Means Clustering

Availability of blood at blood banks in Malang city fluctuates unevenly every month and every year. This greatly affects the demand for blood which is very high every day, especially during the current pandemic. To meet the high demand for blood, a program is needed that aims to attract Malang city residents, especially the productive age segment to donate blood so that the availability of blood every month can be met. This study aims to look at blood stock data from BPS (Central Statistics Agency) Malang City every month for three years (2018-2020) which will be clustered / grouped using K-Means method according to the productive age segmentation (17, 18-24, 25-44, 45-59 years) so that it will produce information when there will be an increase in blood donor raising program according to age segmentation by maximizing the month that is the favorite month of the age segmentation and the month that is quiet for donors. This research resulted in cluster groups based on months with analysis on the best number of clusters so that the evaluation resulted in knowledge that the highest month with many donors was October, while the lowest month with few donors was June and December. The government of Malang and related parties can make a program to increase and maximize the blood donor program in that month.

PENERAPAN ALGORITMA K-MEANS CLUSTERING PADA PENGELOMPOKKAN PASANGAN USIA SUBUR PESERTA KB DI KABUPATEN CIREBON

One of the problems faced by newly married couples is planning the birth of children in the family. The problem of regulating the birth spacing of children in the family can be regulated by the family planning program that has been launched by the Government. One way to suppress the problem of population growth rate in Indonesia is through the KB (Family Planning) program. The success of the family planning program can be seen from the contraceptive methods used by couples of childbearing age (PUS) in each region. The main purpose of this study was to determine the best group of couples of childbearing age for family planning participants based on the contraceptives used in Cirebon Regency using the K-means algorithm, where this algorithm can solve the clustering problem in determining the best group of couples of childbearing age family planning participants. The results of the clustering process to find the best group of couples of childbearing age for family planning participants by looking for the smallest DBI value and obtained K-2 as the best group with a Davies Bouldin Index value of 0.071. The best number of clusters from the grouping of fertile age couples for family planning participants is 2 clusters, cluster 0 has 20 items members and cluster 1 has 20 items members.&#x0D; &#x0D; Keywords: Clustering, K-Means, Family planning(KB), Couples of childbearing age(PUS)

Structured Populations of Critically Endangered Yellow Water Lily (Nuphar shimadai Hayata, Nymphaeaceae).

Yellow water lily (Nuphar shimadai Hayata) is a critically endangered species in Taiwan. Here, we examined genetic structures of four extant populations, WP, GPa, GPb and GPn, using 39 simple sequence repeat (SSR) markers. Positive genetic correlation was observed within 50 m, beyond which no correlation was detected due to isolation by distance according to Mantel correlogram. This suggests a significant genetic structuring of the species. Besides, multilocus genotype (MLG) analysis revealed that GPa was a panmictic population and the species’ putative center of origin. Genetic exchange was observed between GPa and GPb populations, which likely resulted from their geographic proximity. Nevertheless, there was a strong asymmetric migration detected from GPa to WP, but a recent genetic barrier prevented dispersal further northward (WP). Geneland estimated the best number of clusters as K = 2, where WP distinctly separated from the rest of the populations. In STRUCTURE output of K = 3, a third cluster was abundant only in WP. We suggest to consider GPn and WP as separate conservation units, being far from GPa. There is indeed a need to investigate these populations; as predicted, Ne = 1.6 to 3.0 is considered low and that may put the species at risk of extinction.

A deep density based and self-determining clustering approach to label unknown traffic

Analyzing non-labeled data is a major concern in the field of intrusion detection as the attack clusters are continuously evolving which are unknown for the system. Many studies have been conducted on different techniques such as clustering to solve this issue. Consequently, in this paper the clustering techniques are applied based on the packets’ similarity to categorize unknown traffic. After clustering is done by the proposed architecture, the security investigator analyzes one packet from each cluster (instead of thousands of packets) and generalize the result of analysis to all packets belonging to the cluster. The proposed architecture, namely Associated Density Based Clustering (ADBC) applies multiple unsupervised algorithms and a co-association matrix to detect attack clusters of any shape as long as they have density-connected elements. Furthermore, the architecture automatically determines the best number of clusters in order to categorize non-labeled data. The performance of proposed architecture is evaluated based on the various metrics, while its generalization capability is tested with three publicly available datasets.

PENGELOMPOKAN PROVINSI DI INDONESIA BERDASARKAN INDIKATOR KESEHATAN LINGKUNGAN MENGGUNAKAN METODE PARTITIONING AROUND MEDOIDS DENGAN VALIDASI INDEKS INTERNAL

Environmental health is an important aspect in efforts to achieve public health. The condition of environmental health in Indonesia is varies in each province, so the priorities for increasing environmental health are also different. This study aims to grouping provinces in Indonesia based on environmental health indicators in order to know the high/low environmental quality in each province to assist the government in optimizing environmental health efforts. The grouping of provinces is done partitioning around medoids method which is robust to data containing outliers. The measure of similarity objects is calculated using the Euclidean and Manhattan distances, the selection of the best number of clusters is done by validating the internal index, namely the Calinski-Harabasz index, Baker-Hubert index, silhouette index, C-index, and Davies-Bouldin index. The result of this study is that the best number of clusters are two clusters using the Manhattan distance measurement method, with the largest Calinski-Harabasz index value = 24.10072, the largest Baker-Hubert index = 0.8466251, the largest silhouette index = 0.4246581, the smallest C-index = 0.07290109, and the smallest Davies-Bouldin index = 1.094805.

Kernel K-Means Clustering untuk Pengelompokan Sungai di Kota Semarang Berdasarkan Faktor Pencemaran Air

K-Means Clustering is one of the types of non-hierarchical cluster analysis which is frequently used, but has a weakness in processing data with non-linearly separable (do not have clear boundaries) characteristic and overlapping cluster, that is when visually the results of a cluster are between other clusters. The Gaussian Kernel Function in Kernel K-Means Clustering can be used to solve data with non-linearly separable characteristic and overlapping cluster. The difference between Kernel K-Means Clustering and K-Means lies on the input data that have to be plotted in a new dimension using kernel function. The real data used are the data of 47 rivers and 18 indicators of river water pollution from Dinas Lingkungan Hidup (DLH) of Semarang City in the first semester of 2019. The cluster results evaluation is used the Calinski-Harabasz, Silhouette, and Xie-Beni indexes. The goals of this study are to know the step concepts and analysis results of Kernel K-Means Clustering for the grouping of rivers in Semarang City based on water pollution factors. Based on the results of the study, the cluster results evaluation show that the best number of clusters K=4

Differential Evolution Model for Identification of Most Influenced Gene in Brest Cancer Data

Microarray technology generates a large amount of data. Clustering is a popular technique for locating genes that are expressed in close proximity. It entails examining a fresh dataset to determine whether similar traits can be used to identify any hidden groupings. There are large-dimensional datasets accessible, such as those produced from gene expression investigations, RNA microarray studies, or RNA sequencing studies. As a consequence, cluster analysis and producing well-separated clusters become more challenging. Good cluster separation is desirable since it suggests that items are not being placed in the erroneous clusters. In this study, it was recommended that a Differential Evolution-based (DE) Model be used to interpret the analysis of Brest cancer gene expression. To begin, cluster the gene expression data to find the genes most likely to be impacted by the illness. The appropriate number of clusters must be found in order to locate the gene with the highest effect in the gene collection. We used a DE model on the Brest cancer datasets in this work. We identified the best number of clusters by using the most impacted gene in this dataset as a benchmark. We then experimented with different cluster sizes. We used the DE method to three distinct breast cancer datasets and compared it to the current K-Means and K-medoids models. The results of the experiments show that the proposed DE model outperforms existing models significantly.

Optimization of the Number of Clusters of the K-Means Method in Grouping Egg Production Data in Indonesia

The need for eggs that continues to increase will not increase with large egg production so that there is a shortage of egg supplies which results in high egg prices. It is necessary to group egg production in Indonesia to find out which areas fall into the high cluster and which areas fall into the low cluster. This study aims to classify the egg production of laying hens in Indonesia. The method used is the K-Means Clustering method which is a popular clustering method. To find out how optimal the number of clusters in the K-Means method is for grouping egg production in Indonesia, the researcher evaluates the DBI value of each number of existing clusters. In this study, 8 clusters were used, namely 2 clusters, 3 clusters, 4 clusters, 5 clusters, 6 clusters, 7 clusters, 8 clusters, and 9 clusters. The results of measuring the DBI value are the number of clusters 2 = 0.215, the number of clusters 3 = 0.149, the number of clusters 4 = 0.146, the number of clusters 5 = 0.157, the number of clusters 6 = 0.180, the number of clusters 7 = 0.205, the number of clusters 8 = 0.192 and the number of clusters 9 = 0.154. This study shows that the best number of clusters is the number of clusters 4 with the smallest DBI value of 0.146.

Comparative Study of K-Means Clustering Algorithm and K-Medoids Clustering in Student Data Clustering

Universities as educational institutions have very large amounts of academic data which may not be used properly. The data needs to be analyzed to produce information that can map the distribution of students. Student academic data processing utilizes data mining processes using clustering techniques, K-Means and K-Medoids. This study aims to implement and analyze the comparison of which algorithm is more optimal based on the cluster validation test with the Davies Bouldin Index. The data used are academic data of UIN Sunan Kalijaga students in the 2013-2015 batch. In the k-Means process, the best number of clusters is 5 with a DBI value of 0.781. In the k-Medoids process, the best number of clusters is 3 with a DBI value of 0.929. Based on the value of the DBI validation test, the k-Means algorithm is more optimal than the k-Medoids. So that the cluster of students with the highest average GPA of 3,325 is 401 students.