Claim Missing Document
Check
Articles

Found 9 Documents
Search
Journal : Eksponensial

Pengelompokan Kabupaten/Kota Di Pulau Kalimantan Berdasarkan Indikator Indeks Pembangunan Manusia Tahun 2020 Menggunakan Optimasi K-Means Cluster Dengan Principle Component Analysis (PCA) Anwar, Khoiril; Goejantoro, Rito; Prangga, Surya
EKSPONENSIAL Vol. 13 No. 2 (2022)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (964.676 KB) | DOI: 10.30872/eksponensial.v13i2.1053

Abstract

Data mining is a technique or process to obtain useful information from a large database. Based on its functionality, one of the tasks of data mining is to group data. Cluster analysis is an analysis that aims to group objects based on the information found in the data. One of the cluster analysis methods is the K-Means cluster method, which is a non-hierarchical grouping method by dividing the data set into a number of groups that do not overlap between one group and another. This study aims to classify districts/cities on the island of Kalimantan based on indicators of the human development index and obtain the sillhoutte coefficient value from the optimal cluster analysis using the K-Means algorithm on principle component analysis. The data used is the 2020 human development index data in districts / cities on the island of Kalimantan and used 8 variables from the human development index indicator. The results of the optimal cluster formed in the grouping of regencies/cities on the island of Kalimantan using the K-Means cluster method on the principle component analysis are 4 clusters. Cluster 1 has 20 regencies/cities, cluster 2 has 3 regencies/cities, cluster 3 has 26 regencies/cities and cluster 4 has 7 regencies/cities. The sillhoutte coefficient value for data validation from district/city clustering on the island of Kalimantan using the K-Means cluster method on principle component analysis produces 4 clusters of 0.540 which states that the cluster structure formed in this grouping is a medium structure.
Aplikasi K-Nearest Neighbor Dengan Fungsi Jarak Gower Dalam Klasifikasi Kelulusan Mahasiswa: Studi Kasus : Mahasiswa Program Studi Statistika, Jurusan Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam, Universitas Mulawarman Fadil, Irfan; Goejantoro, Rito; Prangga, Surya
EKSPONENSIAL Vol. 13 No. 1 (2022)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (649.085 KB) | DOI: 10.30872/eksponensial.v13i1.881

Abstract

The results of the reaccreditation of the Statistics Study Program, Mulawarman University in 2019 remain accredited B. One of the assessment indicators used in reaccreditation is the student's timely graduation status. Therefore, it is necessary to predict the graduation status of Statistics students, Mulawarman University.. The prediction method used in this research is K-Nearest Neighbor (K-NN). K-NN is a classification method based on studying previously classified data. This method is very easy to understand, easy to applied and also non-parametric method, so that no certain assumptions are needed in the process. The independent variables used in this study were student profiles, including gender, regional origin, cumulative Grade Point Average (GPA) and single tuition fee. The dependent variable in this study is the graduation status of students, namely graduating on time and not graduating on time. The data used were students of the Mulawarman University, Statistics Study Program in 2014, 2015, and 2016. The results showed at k = 7 and the distribution of training and testing data with the proportion of 80:20 obtained optimal accuracy of 0,909 with a TPrate of 0.500, a TNrate. in the amount of 1,000 and AUC value of 0,75 that means fair classification.
Optimasi Fuzzy C-Means Menggunakan Particle Swarm Optimization Untuk Pengelompokan Kabupaten/Kota Di Pulau Kalimantan (Studi Kasus : Data Indikator Kesejahteraan Rakyat Tahun 2020) Febriyanti, Nur Afifah; Goejantoro, Rito; Prangga, Surya
EKSPONENSIAL Vol. 14 No. 1 (2023)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1148.931 KB) | DOI: 10.30872/eksponensial.v14i1.1095

Abstract

Fuzzy C-Means (FCM) is a method of grouping data based on the degree of membership whose observation object is based on the information found in the data describing the object. The FCM method has weaknesses in the initial cluster center determination, so it can be overcome by the Particle Swarm Optimization (PSO) method that can be applied to find the optimal solution of the optimal cluster center determination. The purpose of this research is to determine the optimal number of clusters based on the validity indexes of Partition Coefficient (PC) and Modified Partition Coefficient (MPC), and obtain the results of grouping regencies/cities using the FCMPSO method. Based on the FCMPSO method with a validity index of PC and MPC, it produces an optimal cluster of two clusters, the first cluster consisting of 33 regencies/cities on Kalimantan Island and the second cluster consisting of 23 regencies/cities on Kalimantan Island.
Klasifikasi Status Pembayaran Kredit Barang Elektronik dan Furniture Menggunakan Support Vector Machine Casuarina, Indah Putri; Hayati, Memi Nor; Prangga, Surya
EKSPONENSIAL Vol. 13 No. 1 (2022)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (558.5 KB) | DOI: 10.30872/eksponensial.v13i1.887

Abstract

Classification is the process of finding a model or function that can describe and differentiate data into classes. One application of classification is Support Vector Machine (SVM). SVM is a learning system that uses a hypothetical space in the form of linear functions in a high-dimensional feature space, trained with a learning algorithm based on optimization theory by implementing machine learning derived from statistical learning theory. The concept of classification with SVM is to find the best hyperplane to separate the two data classes and use a support vector approach. This study uses the proportion of the distribution of training data and testing data, namely 50%:50%, 70%:30%, 90%:10% and uses the SVM algorithm Polynomial kernel function with parameters =0.01, r=0.5, d =2, and C=1. This study aims to determine the results of the classification of the credit payment status of electronic goods and furniture and the level of classification accuracy in the SVM method. The data used is the debtor data of PT. KB Finansia Multi Finance Bontang in 2020 as many as 133 data with current and non-current credit payment status and using 7 independent variables, namely age, number of dependents, length of stay, income, years of service, large credit payments, and length of credit borrowing. The results of the SVM classification show an average accuracy value of 72.25% and the best accuracy chosen is the proportion of training data distribution and testing data 90%:10%, which is 84.62%.
Perbandingan Algoritma C4.5 Dan Naïve Bayes Untuk Prediksi Ketepatan Waktu Studi Mahasiswa: Studi Kasus: Program Studi Statistika Universitas Mulawarman Permana, Jordan Nata; Goejantoro, Rito; Prangga, Surya
EKSPONENSIAL Vol. 13 No. 2 (2022)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1043.881 KB) | DOI: 10.30872/eksponensial.v13i2.947

Abstract

Classification is a statistical technique that aims to classify data into classes that already have labels by building a model based on training data. There are many methods that can be used in the classification including Naïve Bayes and C4.5. The C4.5 algorithm is an algorithm used to form a decision tree while Naïve Bayes is a classification based on probability. This study aims to determine the results of the classification of C4.5 and Naïve Bayes and to determine the classification accuracy of the two methods. The variables used in this study were graduation status , entrance , gender , regional origin , GPA , and UKT group . After the analysis, the results showed that the average accuracy level of the C4.5 algorithm was 61.99% and the Naïve Bayes accuracy level was 69.97%. So it can be said that the Naïve Bayes method is a better method in classifying student status compared to the C4.5 . method.
Optimalisasi K-Means Cluster dengan Principal Component Analysis pada Pengelompokan Kabupaten/Kota di Pulau Kalimantan Berdasarkan Indikator Tingkat Pengangguran Terbuka Rais, Muhammad; Goejantoro, Rito; Prangga, Surya
EKSPONENSIAL Vol. 12 No. 2 (2021)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (553.224 KB) | DOI: 10.30872/eksponensial.v12i2.805

Abstract

Data mining or often also called knowledge discovery in databases is an activity that includes collecting, using historical data to find regularity, patterns, or relationships in large data sets resulting in useful new information. Cluster analysis is an analysis that aims to group data based on its likeness. This research uses the K-Means method combined with PCA. The K-Means method groups data in the form of one or more clusters that share the same characteristics. While the PCA method was used to reduce research variables. This grouping method was applied to the data indicator of the unemployment rate of districts/cities in Kalimantan Island in 2018. The cluster validation used in this study was the Davies-Bouldin Index (DBI). Based on the results of the analysis, it was concluded that the number of principal components formed was as many as 2 principal components. The most optimal grouping of districts/cities in Kalimantan island in 2018 was to use 2 clusters with a DBI value of 0,507. The grouping of districts/cities in Kalimantan Island in 2018 produced 2 clusters, cluster 1 consisting of 51 districts/cities and clusters of 2 consisting of 5 districts/cities. Cluster 1 was a cluster that has the highest percentage of the poor population and the highest labor force participation rate when compared to cluster 2. While cluster 2 was a cluster that has an index value of human development, population, number of the labor force, number of unemployed, population density, and the minimum wage of district/city was high compared to cluster 1.
Clustering Titik Panas Bumi Pada Potensi Kebakaran Hutan Menggunakan K-Affinity Propagation Primantoro, Sudhan; Goejantoro, Rito; Prangga, Surya
EKSPONENSIAL Vol. 15 No. 2 (2024): Jurnal Eksponensial
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30872/eksponensial.v15i2.1299

Abstract

K-Affinity Propagation is a development of affinity propagation from Brendan J. Frey and Delbert Dueck. The purpose of this research is to cluster geothermal hotspots on potential forest fires in Indonesia using K-Affinity Propagation for the period July 2022 and obtain optimal cluster results using standard deviation with ratio calculations. The optimal cluster results are 4 clusters, with the number of members in cluster 1 being 12 members with copies in West Sumatera Province, the number of members in cluster 2 being 12 members with copies in Southeast Sulawesi Province, the number of members in cluster 3 being 4 members with copies in Central Sulawesi Province, the number of members in cluster 4 being 1 member with copies in North Sulawesi Province. The optimal cluster results using standard deviation with the smallest ratio value is cluster 4 with a ratio value of 0.057.
Penerapan Algoritma Divisive Analysis dalam Pengelompokan Provinsi di Indonesia Berdasarkan Prevalensi Stunting Suyono, Ari Krisna; Hayati, Memi Nor; Siringoringo, Meiliyani; Prangga, Surya; Fathurahman, M.
EKSPONENSIAL Vol. 15 No. 2 (2024): Jurnal Eksponensial
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30872/eksponensial.v15i2.1341

Abstract

Cluster analysis is an analysis that aims to group data (objects) based only on the information contained in the data that describes objects and the relationships between the objects. Divisive analysis is a clustering method using a top-down approach which starts by placing all objects into one cluster or what is called a hierarchical root and then dividing the cluster root into several smaller clusters. This research aimed to group 34 provinces in Indonesia into 2,3, and 4 clusters based on stunting prevalence data and factors causing stunting in 2022 using a divisive analysis algorithm. The results showed that for 2 clusters, cluster 1 consisted of 32 provinces with low stunting prevalence, and cluster 2 consisted of 2 provinces with high stunting prevalence. For 3 clusters, cluster 1 consisted of 26 provinces with moderate stunting prevalence, cluster 2 consisted of 6 provinces with low stunting prevalence, and cluster 3 consisted of 2 provinces with high stunting prevalence. For 4 clusters, cluster 1 consisted of 21 provinces with moderate stunting prevalence, cluster 2 consisted of 5 provinces with low stunting prevalence, cluster 3 consisted of 6 provinces with high stunting prevalence, and cluster 4 consisted of 2 provinces with very high stunting prevalence.
Prediksi Curah Hujan di Kabupaten Berau Menggunakan Support Vector Regression Patiallo, Monalisa Randang; Fathurahman, M.; Prangga, Surya; Nadhilah Widyaningrum, Erlyne
EKSPONENSIAL Vol. 16 No. 2 (2025): Jurnal Eksponensial
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30872/v16i2.1508

Abstract

Machine learning is an analytical approach that is able to predict the output of a system based on patterns that have been formed from previous data. One of the machine learning methods used in this research is Support Vector Regression (SVR). SVR is the application of the support vector machine method in the case of regression. The concept of the SVR algorithm is to obtain a function with the minimum error rate so as to produce a good predictive value. The advantage of SVR lies in its ability to handle nonlinear data using the kernel functions. This study aims to determine the results of rainfall prediction in Berau Regency using the SVR method. The data used is rainfall data in Berau Regency from January 2014 to December 2023 as much as 120 data, and uses five predictor variables namely temperature, humidity, air pressure, wind speed, and solar irradiation. The kernel function used is a polynomial kernel with parameter values  and . The results showed that the best SVR model to predict rainfall in Berau Regency is the SVR model with parameter values  and . This model provides good prediction performance, with an RMSE value of 0,1786.