Claim Missing Document
Check
Articles

Self Organizing Maps Method for Grouping Provinces in Indonesia Based on the Landslide Impact Suwanda Risky; Syafriandi; Dony Permana; Dina Fitria
UNP Journal of Statistics and Data Science Vol. 1 No. 3 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss3/15

Abstract

Indonesia is a disaster-prone country due to its climatic, soil, hydrological, geological, and geomorphological conditions. A disaster is an event or chain of events that threatens and disrupts people's lives and livelihoods. A natural disaster is a disaster caused by an event or series of events caused by nature such as a landslide. The number of landslide disaster events in Indonesia varies from province to province, this is due to differences in the characteristics of each province in Indonesia. So that the impact caused by the landslide disaster is also different. Therefore, it is necessary to group and profile so that it can be known which province has the largest impact on landslide disasters. This study used the Self Organizing Maps method in a grouping. The number of clusters to be formed is 3 based on the optimal value of internal cluster validation (Dunn, Connectivity, and Silhouette Index). Cluster 1 consists of 31 provinces, and the average impact of landslides is small. In cluster 2 consisting of 2 provinces, there are 4 dominantly more significant impacts. Cluster 3 consisting of 1 province has 1 dominant impact greater. So it can be concluded that most provinces in Indonesia have a relatively small impact on landslide disasters. However, some provinces have a very large impact on landslides, namely the provinces of West Java, Central Java, and East Java.
Comparison of Haversine and Euclidean Distance Formula for Calculating Distance Between Regencies in West Sumatra Vinka Haura Nabilla; Indonesia; Dony Permana; Fadhilah Fitri
UNP Journal of Statistics and Data Science Vol. 1 No. 3 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss3/39

Abstract

A distance is a number that indicates how far apart two place are. The benefits of using distance are widely used in research, one of which is in the application of spatial weighting matrices. The spatial weight matrix is obtained based on proximity information between regions. There are two types of spatial weights, namely, based on contiguity and distance. Determining the proximity of regions in West Sumatra is better to use spatial weighting based on distance because in West Sumatra there are islands and mountains that limit the regions. Some distance estimation equations that can be utilized are Haversine and Euclidean distance. The connection between the two points in Haversine takes into account the earth's curvature when calculating the distance, which is a difference between the two formulas. In contrast, the Euclidean distance method uses a straight line to connect two points. The purpose of this research is to ascertain whether the Haversine and Euclidean distance formulas produce significantly different results in terms of distance. Calculation of the coordinate point distance utilizes latitude and longitude obtained from Google Maps. The distances measured using both formulas were expressed as kilometers (km), then the data was processed using the z test. The findings demonstrated that the Haversine formula and the Euclidean distance formula did not significantly differ in the process of calculating distance.
Prediksi Harga Saham PT Bank Syariah Indonesia Tbk Menggunakan Support Vector Regression Isra Miraltamirus; Fadhilah Fitri; Dodi Vionanda; Dony Permana
UNP Journal of Statistics and Data Science Vol. 1 No. 3 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss3/43

Abstract

A company needs funding from outside the company so that all aspects of development needed can be fulfilled. Companies that need capital can carry out public offerings and sell securities on a stock exchange company. The movement of stock prices tends to fluctuate, so that it will have an impact on the income that will be received by companies and investors. This problem is currently happening to PT BSI Tbk, so it is necessary to do stock price modeling to predict the value of PT BSI Tbk's stock price in the coming days. Support vector regression is a machine learning method that can deal with fluctuating data by producing good predictive models. SVR aims to find the optimal hyperplane to produce a good predictive model. SVR uses the kernel function to handle non-linear data by mapping data from the input space to a higher feature space, hence it will be easier to form an optimal hyperplane. The kernel function used in this study is the radial basis function. The results of this study are that the best parameters are obtained with C = 100, ϵ = 0.01, and γ = 0.001 and produce a model error accuracy of 0.87%.
Classification for Covid-19 Affected Family Cash Aid Recipients Using Naïve Bayes Algorithm Mutiara Amazona Sosiawati; Syafriandi Syafriandi; Dony Permana; Zilrahmi
UNP Journal of Statistics and Data Science Vol. 1 No. 3 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss3/53

Abstract

The COVID-19 pandemic that occurred in Indonesia had a huge impact on the country's economy. One of the solutions set by the government in dealing with COVID-19 is to use APBD funds for social assistance in the form of cash, namely "Village Direct Cash Assistance" (BLT DD). With the hope that the people affected by COVID-19 can be helped by this assistance. There are several problems in the distribution of social assistance, one of which is recipients who are not on target. Therefore, it is necessary to use methods to correctly classify recipients. This study uses the Naïve Bayes method to classify people who receive and do not receive aid. From the results obtained on the confussion matrix, the people who received BLT DD assistance and were predicted to receive were as many as 33 people/KK, the people who did not receive BLT DD and were predicted not to receive as many as 34 people/KK, the people who received BLT DD and were predicted not to receive as many as 2 people/KK , and people who do not receive BLT DD and are predicted to receive as many as 6 people/families. As for the classification accuracy value obtained using the Naïve Bayes method is 89%, while the error rate obtained is 11%.
Modeling Human Development Index in Papua and West Sumatera with Multivariate Adaptive Regression Spline Yulia Pertiwi; Dony Permana; Nonong Amalita; Admi Salma
UNP Journal of Statistics and Data Science Vol. 1 No. 3 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss3/54

Abstract

The Human Development Index (HDI), is an indicator of the successful development of the quality of human life. The high value of HDI, shows the better development of a region. The purpose of this study is to model and determine the factors affect HDI in Papua Province and West Sumatera Province, using Multivariate Adaptive Regression Spline (MARS). MARS is one of the modeling methods that can handle high-dimensional data. The result of this study showed that the best MARS model for Papua Province is a combination of (BF=24, MI=2, and MO=0) with a minimum GCV value of 0.55953. while the best MARS model for West Sumatera Province is a combination of (BF=24, MI=2, and MO=0) with a minimum GCV value of 0.02697. Based on the model, the factors that significantly affect HDI in Papua Province and West Sumatera Province are average years of schooling (X2), adjusted per-capita income (X6), life expectancy (X1), percentage of poor people (X4), and gross regional domestic product (X3). The percentage level of importance of each variable for Papua Province is 100%, 45.26%, 29.24%, 6.55%, and 6.27%. Meanwhile, for West Sumatera Province it is 100%, 96.73%, 57.54%, 34.13%, and 29.6%, respectively. So in this case, based on the results of the study, the average years of schooling (X2) is the variable that most influences HDI in the two regions, with an importance level of 100%.  
Grouping Level of Poverty Based on District/City in Indonesia Using K-Harmonic Means nabillah putri; Nonong Amalita; Dodi Vionanda; Dony Permana
UNP Journal of Statistics and Data Science Vol. 1 No. 3 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss3/60

Abstract

Indonesia still has a relatively high poverty rate, although nationally it has declined in recent years. There are areas that are still experiencing increasing poverty rates. So that the currently planned poverty alleviation plans are no longer uniform, but need to pay attention to the conditions of each dimension that cause poverty in an area, so it is necessary to group districts/cities in Indonesia on poverty. Grouping was performed using K-Harmonic Means analysis. K-Harmonic Means is a non-hierarchical clustering that takes the average of the harmonic distance between each data point and the cluster’s center. The data used in this research is secondary data sourced from BPS publications on poverty and inequality in 2022. The analysis technique is carried out by standardizing the data, conducting cluster analysis, and validating clusters. Based on the results of the K-Harmonic Means analysis, the optimal number of clusters is two clusters that first cluster has 54 districts/cities while second cluster has 460 districts/cities and the Dunn Index value for cluster validation is 0,03492. So that a better grouping level of poverty based on district/city in Indonesia is obtained by using the K-Harmonic Means method with p = 2,25.
Grouping The Regencies/Cities in Indonesia Based on Expenditure Groups Inflation Value Using DBSCAN Method Meliani Putri; Dony Permana; Syafriandi Syafriandi; Zilrahmi
UNP Journal of Statistics and Data Science Vol. 1 No. 3 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss3/61

Abstract

The different characteristics of each regencies/cities in Indonesia can trigger differences in expenditure groups inflation value, the differences that occur will affect Indonesia’s national inflation. The purpose of this research is to create groups of regencies/cities based on expenditure groups inflation value and to identify the characteristics of the resulting groups. DBSCAN is a density-based non-hierarchical cluster method that can be used in data conditions that contain noise. The data used in this study is secondary data obtained from the publication of the Badan Pusat Statistik Republic of Indonesia (BPS RI) regarding expenditure groups inflation value. The analysis includes outlier detection, grouping using the DBSCAN method, performing cluster validation with silhouette coefficient, and identifying the characteristics of the clusters formed. Based on the grouping that has been done, two clusters are produced with a silhouette coefficient value of 0.65. The resulting cluster is cluster 0 in the form of a noise cluster consisting of 3 regencies/cities with regencies/cities that have a high category expenditure groups inflation value. Cluster 1 consisting of 87 regencies/cities is a cluster with regencies/cities that have a low category expenditure groups inflation value.
Pemodelan Waktu Survival Pasien Tuberkulosis menggunakan Regresi Cox Proportional Hazard dengan Data Tersensor Elsa Oktaviani; Nonong Amalita; Atus Amadi Putra; Dony Permana
UNP Journal of Statistics and Data Science Vol. 1 No. 4 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss4/65

Abstract

Tuberculosis is an infectious disease that needs to be watched out for in West Sumatra Province. West Sumatra Province is the province with the 12th highest TB case in Indonesia in 2021 with a total of 8,216 TB cases and a TB treatment cure rate that is still far from the target of the Indonesian Ministry of Health. The purpose of this study is to determine the Cox proportional hazard regression model and factors that affect the survival time of tuberculosis patients at Dr. M. Djamil Padang Hospital. The survival period used is the time when the patient is taking TB treatment at RSUP Dr.  M. Djamil Padang in 2021 until the patient is declared dead. The method used in the Cox Proportional Hazard Regression analysis is the Maximum Partial Likelihood Estimation Method. By using the cox proportional hazard regression model, the factors that influence the survival time of tuberculosis patients at RSUP Dr.  M. Djamil's BMI , leukocytes , fever , shortness of breath , and decreased appetite . 
Sentiment Analysis of Electric Cars Using Naive Bayes Classifier Method NURUL AFIFAH; Dony Permana; Dodi Vionanda; Dina Fitria
UNP Journal of Statistics and Data Science Vol. 1 No. 4 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss4/68

Abstract

In recent years, electric cars have become increasingly popular as an alternative to environmentally friendly vehicles in the automotive industry. These vehicles use electric power as an energy source that can mitigate the reliance on fossil fuels contribute to efforts to minimize greenhouse gas emissions and air pollution. However, the presence of electric cars raises pro and con opinions from the public. the conversation about electric cars has become one of the hot on social media. Twitter is a social media microblogging that permits its users to create short messages and share them easily and quickly. These opinions require sentiment analysis. The purpose of conducting sentiment analysis is to find out how people's perceptions and opinions on electric cars are leading in a favorable or unfavorable direction. Thus, sentiment analysis can help companies marketing strategies, and better business decisions. Then the opinions will be classified based on positive and negative categories. This investigation employs the naive classifier method to generate positive and negative sentiment towards electric cars on Twitter. The accuracy results of naive bayes obtained by using a confusion matrix in this research are 77.8%, with a dataset split composition of 70%:30%.
Analysis of the Poverty Level Model for West Sumatra Province Using Geographically Weighted Binary Logistic Regression april leniati; Dony Permana; Nonong Amalita; Zamahsary Martha
UNP Journal of Statistics and Data Science Vol. 1 No. 4 (2023): UNP Journal of Statistics and Data Science
Publisher : Departemen Statistika Universitas Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24036/ujsds/vol1-iss4/80

Abstract

T   West Sumatra Province (West Sumatra) ranks third lowest in terms of the poverty rate on the island of Sumatra in 2022, with a figure of 5.92%. Although this figure is lower than the national average, the Province of West Sumatra is targeting a reduction in the poverty rate to 5.62% in 2024 in the vision of the 2021–2026 Regional Development Plan. The purpose of this study is to analyze the factors that contribute to the poverty rate in West Sumatra Province based on geography in 2022. The method used to address poverty problems is Geographically Weighted Binary Logistic Regression (GWBLR), which takes geographical influences into account in the analysis. This study uses data on the percentage of poor people (Y) and the influencing factors, namely life expectancy (X1), literacy rate (X2), labor force participation (X3), and economic growth (X4). The results showed that based on the lowest Akaike Information Criterion Corrected (AICc) value, the GWBLR model with a Fixed Gaussian Kernel weight is the best at modeling the problem of poverty in West Sumatra in 2022. According to the model, the life expectancy variable will have a significant impact on the level of poverty in 13 districts and cities in West Sumatra Province in 2022.
Co-Authors 01, Riska Addini, Vidhiya Ade Eriyen Saputri Admi Salma Admi Salma Afdhal, Afdhal Rezeki Afifah Zafirah Ahmad Fauzan Aidillah, Kerin Hagia Alandra, Cindy Resha Aldi Prajela Ali Asmar Andini Diva Luthfiyah april leniati Armiati Arnellis Arnellis Arssita Nur Muharromah Atus Amadi Putra Azma, Meil Sri Dian Bahri Annur Sinaga Bonita Nurul Afifah Carina, Fadhillah Meisya Denny Armelia Dewi Febiyanti Dina Fitria Dina Fitria, Dina Dinul Haq, Asra Dodi Vionanda Dwi Putri Amilia Dwi Ratih Listiani Yusri Dwi Sulistiowati Edwin Musdi Elita Zusti Jamaan Elsa Oktaviani Elvina Catria Emi Suryani Putri Fadhilah Fitri Fadhillah Fitri Fadlan Rafly, Muhammad Fanni Rahma Sari Fauzan Arrahman Febri Ramayanti Fenni Kurnia Mutiya Fishuri, Nufhika Hana Rahma Trifanni Hana Zafirah haniyathul husna Hardi, Afifah Hasna, Hanifa Hefiani Mustika Hasanah Helma Helma Huriati Khaira I Made Arnawa Ibnul farizi, Gilang iin aini fitri Indonesia Irma Surya Anisa Isra Miraltamirus Kamil, Fakhri Kurnia Andrea Diva martha, Ully Martha Media Rosha Meidiani Sandra Meliani Maya Sari Meliani Putri Mohammad Reza febrino Muslimah, Nailul Amani Muthia Sakhdiah Mutiara Amazona Sosiawati nabillah putri Nadya Nadya nazhiroh, hanifah Nilda Yanti Nisa Ulkhairat Asfar Nisa, Farras Luthfyah Nonong Amalita Nur Fadillah, Nur Nurdalia Nurul Afifah Putra, M. Farel Rusde rahmad revi fadillah rama novialdi Refenia Usman Refina Rintani Revina Rahmadani Ridha Fajria rios Riry Sriningsih RIZKIA, DHEA PUTRI Ronald Rinaldo roza maylinda Salsabilla Khairani Septrina Kiki Arisandi Siltima Wiska Siregar, Fauzan Al-Hamdani Sofni Fajriani SRI RAHAYU Suherman Suherman Suwanda Risky Syafriandi Syafriandi Syafriandi Tessy Octavia Mukhti Titin Mardianingsih Tri Wahyuni Nurmulyati Vinka Haura Nabilla Wahda Aulia Assara Welgi Okta Irawan Widia Handa Riska Yarman Yarman Yatri Asri Yenni Kurniawati Yerizon Yerizon Yoga Perdana Yuli Andari Wulan Yulia Pertiwi Yulia Utami Putri Yulyanti Harisman Yurivo Rianda Saputra YUSWITA, AULIA Zamahsary Martha Zilrahmi, Zilrahmi