Silhouette Coefficient Value Research Articles

Penerapan Metode K-Medoids Clustering Untuk Mengelompokkan Ketahanan Pangan

Food is a basic need that must be fulfilled and easily accessible to the entire community. After the end of the pandemic period, it still caused several sectors to decline, including the agricultural sector, which resulted in crop yields also declining. The problem faced by several regions in Indonesia, one of which is the North Sumatra region, is that the availability of food products has decreased and increased unstably due to the lack of information about the grouping food security every year. This results in the food needs of the people in each region being unfulfilled. The purpose of this study is to group areas with the number of increases and decreases in food crop yields in North Sumatra using the K-Medoids algorithm. The K-Medoids algorithm includes a deflection algorithm that is quite efficient in carrying out the shaking of small datasets and the search for the most representative points and can overcome outliers. So that it can be used in the floundering of the influence of productivity and the level of food security. The results showed that the application of the K-Medoids algorithm resulted in a DBI (Davies Bouldin Index) value of 0.062 and a Silhouette Coefficient value of 0.8980, with the number of clusters as many as 3 clusters where Cluster_0 dominated by corn food crops experienced an increase in production by 5% and peanuts by 5%, Cluster _1 was dominated by a decrease in the number of soybean production yields by 38%, and Cluster_2 dominated by a decrease in green bean yield by 33%.

Research on Personalized Recommendation Method of Educational Resources Based on Learner Behavior Analysis

The explosion of online learning resources makes the research on personalized recommendation of educational resources increasingly prominent. Based on the theory of learner behavior analysis, this paper analyzes the learning behavior logs of the online learning platform, and constructs a personalized recommendation method for educational resources. It analyzes the learner behavior from the three dimensions of basic attributes, behavior characteristics and result characteristics, and solves the problem of resource analysis. In the simulation process, the method uses the vector space model to complete the modeling of learner behavior, and realizes the division of learner groups based on learner behavior, and further evaluates and optimizes the division of learner groups. In order to further verify the clustering effect in combination with the actual data, the classification results are used for example analysis: the experimental results show that the sample points in the educational resource space are divided into 23 categories according to the distance relationship, and [Formula: see text] is the best. The silhouette coefficient value is 0.54436, and the Calinski–Harabasz score value is 1464.9, which effectively improves the clustering effect of the personalized recommendation method.

Pengelompokan Pelanggan Toko Kerajinan Menggunakan K-Means dengan Model RFM dan LRFM

Customer groups within a company or shop are important to identify for determine the right sales strategy in a competitive market. This research was conducted at one of the online handicraft shops in Bali which was affected by the intense competition in the market. The handicraft industry in Bali is considered to provide a large contribution to the community's economy. One solution that can be done is to identify the customer groups in the store using the clustering technique. This research aims to get the best customer data cluster and model. The stages in this research start from preprocessing data on historical data and customer orders to generate data model Recency, Frequency, Monetary (RFM) and Length, Recency, Frequency, Monetary (LRFM), then the clustering process with K-Means and evaluation of cluster quality with Silhouette Coefficient (SC). Results Based on the research, the RFM model becomes a data model that produces the best clustering results with an SC value of 0.545 when k = 2 which is included in the medium structure. LRFM model only produces the largest SC value when k = 3 with a value of 0.415. The SC value calculated from each data model tends to increase when the percentage of data is added. Cluster 1 has 817 members with the last transaction taking a long time, but has a below average ordering and monetary frequency. Cluster 2 has 158 members, making the last transaction at the most recent time, and has an average order and monetary frequency.
 
 Keywords: Clustering, Customer, Craft, K-Means, RFM, LRFM, Silhouette Coefficient

Pemetaan Terhadap Penyebaran Unit Cabang Kantor Pos Di Kabupaten Wonogiri

The Post Office is one of the public facilities for the community because it serves service facilities in the form of postal services, package delivery, logistics and reliable financial services. However, the lack of information makes access to the location of post office services not informed of the location distribution. The purpose of this research is to create a mapping application for branch post office units to make it easier for the general public to get information about the nearest post office unit, using the K-Means Clustering Method. Users can choose which post office unit they want to choose based on their criteria. In functional testing (Blackbox) it was found that the system had been running according to the system design plan that had been made previously, while the validity test using the Silhouette Coefficient method showed the average value for cluster 1 was 0.282674, cluster 2 was 0.297377, and cluster 3 is 0.423516, which means the object is in a good or appropriate group because the Silhouette Coefficient value in each cluster is closer to 1, which means the better the grouping of data in one cluster.Keywords : K-Means Clustering, Post Office, Silhouette Coefficient

Penentuan Pencemaran Air Menggunakan Metode Self Organizing Maps (SOM)

Water pollution is the entry of a component or substance into water, which causes water quality to decrease. The decrease in water quality causes the water to be unfit and affects the community's health. Therefore, an application for determining the water quality level building using the SOM algorithm with several data input parameters: Fluoride, Hardness, and Nitrite. These parameters use for data clustering in determining clean and polluted water. The SOM algorithm arranges SOM neurons based on the data input values ​​in a SOM cluster. For testing the SOM method on 114 data, the Silhouette Coefficient value was used to find the best number of clusters. Silhouette Coefficient will evaluate clustering the proximity of the distance between data. The test was carried out 27 times with variations of the experiment with the learning rate starting from 0.1 to 0.9 and the number of clusters from 2 to 4 to get the best Silhouette Coefficient value. The result of clustering the best silhouette coefficient value obtained is 0.7276473444141 with 3 clusters and the learning rate of 0.2.

Society's Perspectives on Contemporary Islamic Law in Indonesia through Social Media Analysis Technology: A Preliminary Study

The social, cultural, and technological developments of society are unavoidable. This has an impact on the development of Islamic Law, which keeps all Muslim activities in the right corridor. Contemporary Islamic law, known as Contemporary Islamic Law, has also developed to answer new societal problems. Various views on Contemporary Islamic Law in solving multiple issues certainly reap various responses from the community and scholars. These views are often conveyed through social media such as Youtube, Instagram, Facebook, and Twitter. Therefore, this article aims to discuss a preliminary study of text analysis techniques used to find out the views of the community and Ulama on Contemporary Islamic Law issues computationally and automatically. This initial study reviews the methods and techniques that will be used, namely the Indonesian National Work Competency Standards (SKKNI) methodology for data science. This study will also use a sentiment analysis approach, topic modeling, and pattern analysis to find out people's views on issues of Contemporary Islamic Law through social media. The algorithm used for sentiment analysis is the Multinomial Naïve Bayes Classifier (MNBC), for topic modeling is Latent Dirichlet Allocation (LDA), while for pattern analysis using the Prefix-projected Sequential Pattern Mining (PrefixSpan) algorithm. The model generated from sentiment analysis, topic modeling, and pattern analysis will be evaluated by measuring the confusion matrix, coherence value, and silhouette coefficient value. In addition, analysis and interpretation of the model results will be carried out in-depth qualitatively by involving the views and thoughts of Islamic Law experts.

Comparison of clustering in tuberculosis using fuzzy c-means and k-means methods

Tuberculosis (TB) is a health problem that has yet to be resolved in Indonesia. Based on WHO data, in 2021 Indonesia will still be in the third rank of the highest TB cases in the world. This study aims to determine how many groups of TB patients are based on age, gender, HIV status, history of diabetes mellitus, chest X-ray, and the results of the Molecular Rapid Test (TCM). The data used in this study were 985 from 2017 to 2020. The method used in this research is K-Nearest Neighbor (KNN) in carrying out the imputation process, as well as comparing the k-means and Fuzzy C-Means (FCM) methods in classifying TB data. Before doing the grouping, the data cleaning process is carried out by an imputation process which is useful for filling in the missing data in this case, using the KNN method. To produce maximum results of data grouping or clustering, it is necessary to determine the right number of clusters. For this reason, this study tries to compare the elbow, silhouette coefficient, and Davies Bouldin Index (DBI) methods. The application of the KNN method in the data imputation process in this study is to use k=5. The application of the K-Means algorithm is to form groups of TB patients based on six features. Determination of the optimal number of clusters using the K-means and FCM methods shows the optimal number of clusters, namely K = 2 but with different values. The results of the clustering test using the elbow method with the K-means and FCM methods are 93288.49. The DBI value for the K-means and FCM methods is 0.4937. Meanwhile, the clustering trial with the silhouette coefficient on K-means yields a value of 0.6318 which is better than the FCM which produces a value of 0.6321. This shows that the results of clustering k-means with silhouette coefficients produce better cluster quality because they have a lower silhouette coefficient value than FCM.

A combination of algorithm agglomerative hierarchical cluster (AHC) and K-means for clustering tourism in Madura-Indonesia

The development approach through the tourism sector is one of the programs launched by the government since 2016. However, the development approach is not carried out in all areas because the number of accommodation and public facilities is minimal and uneven, one of which is in Madura. With so many tourist objects in Madura, it is necessary to distribute the development of public facilities and analyze tourism that has a non-strategic distance to public facilities to help increase tourist visits. This study builds a system for clustering tourist attractions in each district in Madura based on the distance to public facilities which include hotels, gas stations, restaurants, and mosques which are important criteria and considerations for tourists in visiting a tourist location. The method used in this research is a combination of the AHC method with K-Means. The test results of the AHC, K-Means method, and the combination of AHC and K-Means methods using the Silhouette Coefficient method indicate that the AHC and K-Means combination method is the best method with a Silhouette Coefficient value of 0.8055 for k=2 and is classified as a strong structure, for the K method. -Means produces the highest Silhouette Coefficient value of 0.638. While the AHC method produces the highest Silhouette Coefficient value of 0.707.

PENGELOMPOKAN DESA ATAU KELURAHAN DI KUTAI KARTANEGARA MENGGUNAKAN ALGORITMA DIVISIVE ANALYSIS

Potential Villages (PODES) provide data on the existence, availability and development of the potential of each government administrative area. In order to make it easier for governments to make policies for a region, it is necessary to group the village and sub-districts. Cluster analysis is an analysis that aims to group objects based on the information that found in the data. One of the cluster analysis methods is the divisive analysis, which is a hierarchical grouping method with a top-down approach, where all objects are placed in one cluster and then sequentially divided into separate groups. This research aim to group villages or sub-districts in Kutai Kartanegara based on the determinants of village backwardness and obtaining the silhouette coefficient value from the optimal cluster analysis using the divisive analysis algorithm. The data used is the 2018 PODES data in Kutai Kartanegara and used 15 variables from natural and environmental factors, facilities infrastructure and access factors as well as socio-economic factors of the population. The results of the optimal cluster formed in the grouping of villages or sub-districts in Kutai Kartanegara using the divisive analysis method are 2 clusters. Cluster 1 consisting of 230 villages or sub-districts and cluster 2 consisting of 2 sub-districts. Silhouette coefficient value for data validation from clustering village or sub-districts in Kutai Kartanegara using the divisive analysis method produces 2 clusters is 0,744 which states that the cluster structure formed in this grouping is a strong structure.

ANALISIS KECENDERUNGAN LAPORAN MASYARAKAT PADA “LAPORGUB..!” PROVINSI JAWA TENGAH MENGGUNAKAN TEXT MINING DENGAN FUZZY C-MEANS CLUSTERING

Effective communication between the government and society is essential to achieve good governance. The government makes an effort to provide a means of public complaints through an online aspiration and complaint service called “LaporGub..!”. To group incoming reports easier, the topic of the report is searched by using clustering. Text Mining is used to convert text data into numeric data so that it can be processed further. Clustering is classified as soft clustering (fuzzy) and hard clustering. Hard clustering will divide data into clusters strictly without any overlapping membership with other clusters. Soft clustering can enter data into several clusters with a certain degree of membership value. Different membership values make fuzzy grouping have more natural results than hard clustering because objects at the boundary between several classes are not forced to fully fit into one class but each object is assigned a degree of membership. Fuzzy c-means has an advantage in terms of having a more precise placement of the cluster center compared to other cluster methods, by improving the cluster center repeatedly. The formation of the best number of clusters is seen based on the maximum silhouette coefficient. Wordcloud is used to determine the dominant topic in each cluster. Word cloud is a form of text data visualization. The results show that the maximum silhouette coefficient value for fuzzy c-means clustering is shown by the three clusters. The first cluster produces a word cloud regarding road conditions as many as 449 reports, the second cluster produces a word cloud regarding covid assistance as many as 964 reports, and the third cluster produces a word cloud regarding farmers fertilizers as many as 176 reports. The topic of the report regarding covid assistance is the cluster with the most number of members.

Penggerombolan Sekolah pada Penerimaan Mahasiswa Baru Jalur SNMPTN di IPB Menggunakan Metode Two-Step Cluster

New student admissions are opened in three pathways including SNMPTN, SBMPTN, and Seleksi Mandiri. In order to improve the SNMPTN selection system at IPB, a study was conducted on the quality of SMA/MA which registered to IPB through school clustering. In general, cluster analysis cannot handle large and mixed-type data, so this school clustering used the Two-Step Cluster method with two alternatives, namely without handling outliers and handling 5 percent outliers. Both of these alternatives produced an average Silhouette coefficient value of 0.2 and 0.3 respectively, which was still under the good category. However, clustering without handling outliers resulted in more detailed cluster criteria with 4 optimal clusters. The criteria for these four clusters include, Cluster 1 is a category of Low Commitment, Low Quality, and Low Consistency schools, Cluster 2 and 3 are categories of schools that have special criteria in certain categories, and Cluster 4 is a category of High Commitment, High Quality, and High Consistency.

Metode Fuzzy Subtractive Clustering Dalam Pengelompokkan Penggunaan Energi Listrik Rumah Tangga

The use of electricity in household sector has increased, especially during the Covid-19 pandemic. The large number of activities carried out in home such as Work from Home, online schools, and online businesses caused difficulty to monitor the electricity consumption. The absence of electricity usage provisions affects the electricity monitoring process. Hence it takes a real time monitoring application of electricity consumption. Fuzzy subtractive clustering is an unsupervised method to form the number and center of clusters according to data conditions. This method serves to classify the household electricity users with the parameters used, is the amount of usage in rupiah and electric power. The grouping results from this method help users to monitoring electricity consumption in real time. The output describes the level of high, medium and low user electricity consumption. Based on the test results, the best Silhouette Coefficient value is 0.8322535 and three clusters are formed, with an accept ratio is 0.5, a reject ratio of 0.15, a radius of 1.7 and a squash factor of 0.5 hence a high level of use is obtained with an average value of the number of uses in IDR 655,993, power 2757 VA, medium level 240,553, 1071 VA and low level 46,479, 675 VA

ClusterMix K-Prototypes Algorithm to Capture Variable Characteristics of Patient Mortality With Heart Failure

Cardiovascular Disease (CVD) is one of the leading causes of many death worldwide, leading to heart failure incidence. The World Health Organization (WHO) says the number of people dying from cardiovascular disease from heart failure each year has an average of 17,9 million deaths each year, about 31 percent of the total deaths globally. Identify the mortality factors of heart failure patients that need to be formed, which reduces death due to heart failure. One of them is by using variable mortality due to heart failure by applying the k-prototypes algorithm. The clustering result is formed 2 clusters that are considered optimal based on the highest silhouette coefficient value of 0,5777. The results of the study were carried out as segmentation of patients with variable mortality of heart failure patients, which showed that cluster 1 is a cluster of patients who have a low risk of the chance of mortality due to heart failure and cluster 2 is a cluster of patients with a high risk of mortality due to heart failure. The segmentation is based on the average value of each variable of heart failure mortality factor in each cluster compared to normal conditions in serum creatine variables, ejection fraction, age, serum sodium, blood pressure, anemia, creatinine phosphokinase, platelets, smoking, gender, and diabetes.

Identifying Mitochondrial-Related Genes NDUFA10 and NDUFV2 as Prognostic Markers for Prostate Cancer through Biclustering.

Prostate cancer is currently associated with higher morbidity and mortality in men in the United States and Western Europe, so it is important to identify genes that regulate prostate cancer. The high-dimension gene expression profile impedes the discovery of biclusters which are of great significance to the identification of the basic cellular processes controlled by multiple genes and the identification of large-scale unknown effects hidden in the data. We applied the biclustering method MCbiclust to explore large biclusters in the TCGA cohort through a large number of iterations. Two biclusters were found with the highest silhouette coefficient value. The expression patterns of one bicluster are highly similar to those found by the gene expression profile of the known androgen-regulated genes. Further gene set enrichment revealed that mitochondrial function-related genes were negatively correlated with AR regulation-related genes. Then, we performed differential analysis, AR binding site analysis, and survival analysis on the core genes with high phenotypic contribution. Among the core genes, NDUFA10 showed a low expression value in cancer patients across different expression profiles, while NDUFV2 showed a high expression value in cancer patients. Survival analysis of NDUFA10 and NDUFV2 demonstrated that both genes were unfavorable prognostic markers.

Application of Density Based Spatial Clustering Application With Noise (DBSCAN) in Determining the Quality of Keprok Orange and Siam Orange Hybrid in the Research Center of Orange and Subtropic Plants Batu City

Abstract. One of the tasks of the Indonesian Citrus and Subtropical Research Institute is research on crossing between citrus varieties to produce saplings with the best quality products through observation of the fruit produced. Because the amount of fruit production studied is very large, it requires a fast and accurate observation process, one of which is the clustering method of data mining. Observations were made using a clustering process or grouping Density Based Spatial Clustering Application with Noise (DBSCAN) on fruit characteristics that indicate quality. DBSCAN works by grouping data based on density, so that it is expected to find several data groups that are close to each other which shows the tendency of the quality of the observed fruit data as well as labeling outlays for data that are too far from the crowd. The results of the grouping will be analyzed to find out the number and characteristics of the groups formed where the results of the grouping are assessed using the Silhouette Coefficient method to determine the best parameter values. The results obtained in this study are obtained three group results which will be divided into medium quality, good, and not so good. The quality of grouping using the Silhouette Coefficient value of 0.69.

Analisis strategi dan klasterisasi ketahanan pangan nasional dalam menghadapi pandemi covid-19

Covid-19 outbreak that scours the world nowadays is affecting all sectors, including food security. Therefore it needs to restructuring the food security policies to ensure that every people obtains adequate and nutritious food. However, the society in each province have different conditions. Thus the clusterization of food security level per province is indispensable to support strategic and policy decision in order to face the Covid-19 pandemic. This research aimed to clustering food security level of each province in Indonesia. Furthernore, this research also compare several clustering methods. The clustering method that used as a comparison in this study is K-means, DBSCAN, Louvain and Self organizing maps methods. Method with the highest silhouette coefficient value in this research will represent the results of food security clustering. The resul of the research show that K-means achieve highest silhouette coefficient value (0.568). Therefore the clusterization result of K-means chosen to represent the level of food security in this research. Further, it followed by self organizing maps with silhouette coefficient 0.559, louvain 0.312 and DBSCAN 0.15. The clusterization result show there are 7 provinces with high food security index, 24 provinces with medium food security index and 3 provinces with low food security index. This research also propose policies strategy and recommendation related to regional food security condition in order to face the Covid-19 pandemic. This research is expected to be a consideration of Indonesian government in making policies on national food security.

Penerapan Algoritma K-Medoids pada Pengelompokan Wilayah Desa atau Kelurahan di Kabupaten Kutai Kartanegara

Kutai Kartanegara Regency (Kukar) was recorded as the largest contributor to the poor population in East Kalimantan (Kaltim) Province in 2017, so that appropriate strategies are needed to solve proverty problems. The development strategy is prioritized for the regions with the largest number of poor people. Identification is conducted based on facilities, infrastructures, access, social, population and economy is provided in the Village Potential data (PODES). K-Medoids is a grouping method that uses representative objects as a central point, which can be used to find out the characteristics of a region. This research is aimed to find out the optimal cluster formed by choosing the largest value of Silhouette Coefficient (SC) from the grouping of villages / political district in Kukar Regency using PODES data in 2018. Clusters that will be formed in this research are 2 clusters, 3 clusters, 4 clusters and 5 clusters. Based on the analysis, it can be seen that the value of SC 2 cluster is 0.430, the value of SC 3 cluster is 0.174, the value of SC 4 cluster is 0.175 and the value of SC 5 cluster is 0.196. So that the largest SC or optimal cluster values ​​obtained in the grouping of 2 clusters with a SC value of 0.430. Cluster 1 consists of 186 villages / political dsitrict and cluster 2 consists of 46 villages / political district.

Implementasi Algoritma Clustering Partitioning Around Medoid (PAM) dalam Clustering Virus MERS-Cov

<p><em>Th</em><em>e Middle East Respiratory Coronavirus (MERS-CoV) is a disease caused by a coronavirus. This virus is contagious, but its transmission is not as easy as the common cold, MERS-CoV virus is better susceptible to transmitting through direct contact, for example in people who care about the MERS-CoV virus without the need for virus protection. To determine the characteristics, the MERS-CoV disease virus can be identified by identifying DNA (deoxyribonucleic acid). One technique in understanding the characteristics of life is by grouping. Grouping can be done by grouping DNA into groups that have attributes and functions. The Clustering algorithm used in this study is Partitioning Around Medoid (PAM). This algorithm has the advantage that the results of the grouping process are not by following the order of entering the dataset and overcome sensitivity to noise and outliers. The purpose of this study is to implement the Partitioning Around Medoid (PAM) clustering algorithm in clustering the MERS-CoV virus. This research was conducted through a quantitative descriptive literature study. The implementation of the PAM algorithm on the MERS-CoV DNA sequence obtained 2 clusters with the highest silhouette coefficient value on the number of clusters 2, namely 0.61534. The number of members in Cluster 1 is 84 MERS-CoV DNA sequences and the number of members in Cluster 2 is 16 MERS-CoV DNA sequences.</em></p>

Pemodelan Cluster Loyalitas Customer Menggunakan Algoritma K-Means Dengan Parameter LRIFMQ

Loyal customers are one of the factors that determine the development of a business. Therefore, businesses need a strategy to keep customers loyal, even making customers who were previously less loyal to become more loyal. The strategy used must be right on target according to customer segmentation. The purpose of this paper is to model a cluster of customer loyalty to help businesses in making the right decisions of marketing strategy. Segmentation is done using the k-means algorithm with LRIFMQ (length, recency, interval, frequency, monetary, quantity) as parameters, and the CLV (customer lifetime value) of each cluster is calculated. Data obtained from PT. XYZ (a company engaged in food processing) for one year (1 January 2019 - 31 December 2019), with 337.739 transactions, and 26.683 customers. AHP (analytical hierarchy process) method is used for LRIFMQ weighting because this method has a consistency index calculation. The silhouette coefficient is used to calculate the cluster quality and determine the optimal number of clusters. The best results are obtained with the silhouette coefficient value of 0,632904 with the number of clusters 6.

Comparison Clustering Performance Based on Moodle Log Mining

The use of digital data now saves a lot of information, but still raw and not in the form of knowledge that can be directly seen, therefore we need a data processing to get useful particular understanding from raw data. Data Mining has an important role to process and find useful information from data. This process is also called Knowledge Discovery. Educational Data Mining is a Knowledge Discovery process in the world of education using data mining techniques. One method used in data mining is clustering. By using clustering analysis techniques, data can be grouped into groups without the need for prior knowledge (previous knowledge), so that the data is grouped based on similarity of patterns. This paper compares the K-means, Hierarchical and Louvain clustering methods to see the most appropriate clustering technique in analyzing log activity data in Moodle Learning Management System (LMS). The results of clustering are measured using the Silhouette Coefficient, and then we compare the values and distribution between clusters. In conclusion, Hierarchical clustering produces the highest Silhouette Coefficient value and also this algorithm can detect outlier data as new cluster. Louvain clustering perform very well to find cluster groups in new dataset as the algorithm does not required the number of clusters to be specified before. Louvain clustering can divide more evenly and precision compare to K-means and hierarchical cluster techniques.