Garuda - Garba Rujukan Digital

Perbandingan Kinerja Metode Klasifikasi Chi-square Automatic Interaction Detection (CHAID) dengan Metode Klasifikasi Algoritma C4.5 pada Studi Kasus : Penderita Diabetes Melitus Tipe 2 Di Samarinda Tahun 2015 Muhammad Faisal; Yuki Novia Nasution; Fidia Deny Tisna Amijaya
EKSPONENSIAL Vol 8 No 2 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (712.426 KB)

C4.5 algorithm is tree classification where tree branches can be more than two. In C4.5 algorithm, the decision tree is based on entropy and gain criterias. Chi-Squared Automatic Interaction Detection (CHAID) classification method is a methods which is used to divide data to become a smaller groups based on categorical dependent and independent variables. The purpose of this research is to determine the classification process by C4.5 algorithm and CHAID method for DM type 2 patients. Risk factors for diabetes type 2 are Decline, Age, Gender, Status of Obesity, Diet, and Sports Activity based on the availability of source data from the Hospital of Abdul Wahab Sjahranie Samarinda. The results show that factors which significantly affect the DM type 2 patients are Obesity and Sport Activity. While by using CHAID, the factors which significantly affect the patients are Decline, Obesity, Diet and Sports Activity. The Classification result accuracy of the C4.5 algorithm is 90% and 94% for CHAID method.

Penggunaan Metode Seven New Quality Tools dan Metode DMAIC Six Sigma Pada Penerapan Pengendalian Kualitas Produk Yurin Febria Suci; Yuki Novia Nasution; Nanda Arista Rizki
EKSPONENSIAL Vol 8 No 1 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (243.816 KB)

Product quality control is a technique and activities or planned actions undertaken to achieve, maintain, and improve the quality of products and services to meet with customers standards and satisfaction. This study aim to address the product quality at a company using statistical methods of products control. The methods are Seven New Quality Tools and DMAIC Six Sigma which are used on a product with a brand of Roti Durian Panglima, produced by PT. Panglima Roqiiqu Group in June 2016. Based on the result by using Seven New Quality Tools method, there are five factors that caused defect on Roti Durian Panglima product, which are : human factor, materials, environmental, machine, and work method, which makes the priority of the product improvement lays on human factor. Meanwhile, the use of DMAIC Six Sigma method has obtained performance baseline values at 4,48 Sigma with four kinds of defects on Roti Durian Panglima products, and based on improvement phase using PFMEA method, the priority on product improvement also lays on human factor.

Penerapan Metode Analisis Regresi Logistik Biner Dan Classification And Regression Tree (CART) Pada Faktor yang Mempengaruhi Lama Masa Studi Mahasiswa Chairunnisa Chairunnisa; Yuki Novia Nasution; Ika Purnamasari
EKSPONENSIAL Vol 8 No 2 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (817.808 KB)

Binary Logistic Regression is one of the logistic regression analysis which is used to analyze the relationship between a dichotomous dependent variable with several independent variables. Classification and Regression Tree (CART) is one of the methods that developed to perform classification analysis on dependent variables either on nominal, ordinal, or continuous scale. In this research, Binary Logistic Regression method and Classification and Regression Tree (CART) applied to the data of the students at Faculty of Math and Natural Science Mulawarman University graduated in year 2016 to determine the characteristic of student which is classified according to two categories that is the study period less than or equal to 5 years and study period more than 5 Years, with five independent variables namely GPA Graduates (X1), Gender (X2), Type of Junior School (X3), Domicile (X4), and Major (X5). Factors that influence the study period of the students based on Binary Logistic Regression method are GPA, Gender, Secondary School Type and Major. The result of classification by using CART method is the student who have the study period less than or equal to 5 Years is a student from Chemistry major or have GPA between 3,51 and 4,00, while the study period more than 5 Year is the student who have GPA between 2,00 and 2,75; 2,76 and 3,50. In terms of classification accuracy, Binary Logistic Regression method was able to accurately predict the observation as much as 75.0%, while the CART method was able to accurately predict the observation as much as 77.27%.

Penggunaan Metode Nonparametrik Untuk Membandingkan Fungsi Survival Pada Uji Gehan, Cox Mantel, Logrank, Dan Cox F Fitriani Fitriani; Sri Wahyuningsih; Yuki Novia Nasution
EKSPONENSIAL Vol 7 No 2 (2016)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (390.866 KB)

Survival analysis is a statistical method that aims to study and model the relationship between risk factors and the study time students to reach graduation. In this study conducted a survival analysis using a nonparametric method. They are Gehan Test, Cox Mantel Test, Logrank Test, and Cox F Test on data of students of Mulawarman University Faculty of Mathematical and Natural Science majoring in Statistics and majoring in Computer Science 2010. The purpose of this research was to compare the of period of study survival function students majoring in Statistics and majoring in Computer Science . This study was conducted using data of 167 students majoring in Statistics and majoring in Computer Science. The results showed that students of majoring in Computer Science longer in studying compared with students majoring in Statistics. For students majoring in Statistics who participated in the selection to go to college through the SBMPTN and SMMPTN study longer than SNMPT. While those who while majoring in Computer Sciences who participated in the selection to go to college through three pathways had the same study time.

Analisis Data Kejadian Berulang Tidak Identik Dengan Cox Gap TimeModel Andi Widya Rhezky Awalul Aziz; Yuki Novia Nasution; Sri Wahyuningsih
EKSPONENSIAL Vol 8 No 2 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (987.342 KB)

The gap time method is a method that can be used in recurrent event Time-based modelling. Gap analysis is often useful when events are relatively uncommon, when the object of the study is the prediction of time for the next event, or on the phenomenon of circulation.The analysis of model for non-identical recurrent events using survival time in the form of gap time is called Cox Gap Time Model. The purpose of this research is to know Cox Gap Time model for recurrent occurrence in DM type II disease and to know the factors that influence repetitive incident in DM type II disease in RSUD A. W. Sjahranie Samarinda. The variables in this research are age, treatment, status and relapse time (gap time). The study was conducted by using 263 medical records data of DM type II patients admitted to the hospital during observation period in January 2015 until December 2016. The results shows that age factor affects the first gap time and there are age, gap 1 covariate and gap 2 covariate that have significant effect aga inst to the third gap time variable, meanwhile there is no variable affects the second gap time.

Bootstrap Aggregating Multivariate Adaptive Regression Splines Marisa Nanda Rahmaniah; Yuki Novia Nasution; Ika Purnamasari
EKSPONENSIAL Vol 7 No 2 (2016)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (132.495 KB)

MARS is one of the classification methods that focus on the high dimension and discontinuity of the data. The level of accuracy in MARS can be improved by using Bagging method (Bootstrap Agregating). This method is used to improve stability, accuracy and strength’s of prediction. This study discusses the MARS bagging applications in analyzing the issue of accreditation, which the accreditation level of a schools can be predicted based on the identifier components. Therefore, in this study will be identified these components to create a classification model. The data used is the accreditation data of the primary school in East Kalimantan Province 2015 issued by the Accreditation Board of the Provincial Schools (BAP-S/M) of East Kalimantan Province. This study obtained six components that affect the determination of the accreditation of schools at primary school level. The components are the variables that contribute to the classification. The variables are a standard component of content (X1), a standard component of the process (X2), a standard component of graduates (X3), standard components of teachers and staffs (X4), a standard component of infrastructure (X5) and standard component of financial (X7). Based on the result of the classification accuracy of MARS method (using Apparent Error Rate (APER), it is amounted to 78.87%, while the classification accuracy (using APER) with method of bagging of the best MARS models amounted to 89.44%. This means that the method of bagging MARS gives better classification accuracy of the classification than MARS.

Analysis of (M/G/c): (GD/∞/∞) Menggunakan Software Lazarus Akbar Maulana; Yuki Novia Nasution; Nanda Arista Rizki
EKSPONENSIAL Vol 8 No 2 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (972.023 KB)

Queueing theory is a theory that concerns in the mathematical study of queue or row of waiting. The formation of the queue is occurs when the need for a service exceed the capacity of the service. In this study, an analysis is done to determine whether the queue model (M/G/c) :(GD/∞/∞) can be applied to the Workshop of Utomo Motor Yamaha Samarinda Seberang. Primary data is used and is taken for 3 days in a random busy period selected in May to August 2017. The result is the queue system on Utomo Motor Yamaha using FCFS queue discipline with 4 parallel mechanics, and follows the (M/G/4) :(GD/∞/∞) model. The average of the waiting time in the queue on Monday is 0,38 hours and Wednesday is 0,35 hours. The average of the customers in the queue on Monday and Wednesday is the same as much 2 customer. The average of the customers in the system on Monday and Wednesday is the sama as 5 customers. The average of the waiting times that customers spend on the system on Monday is 0,93 hour and Wednesdayis 0,94 hours and on May 22nd, 2017 is 0,81 hours. In order to calculate the queue model more quickly, a program is made using Lazarus software to search the queue model on daily data.

Aplikasi Data Mining Market Basket Analysis untuk Menemukan Pola Pembelian di Toko Metro Utama Balikpapan Nadya Rahmawati; Yuki Novia Nasution; Fidia Deny Tisna Amijaya
EKSPONENSIAL Vol 8 No 1 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (179.701 KB)

The development of information technology in the transaction process in supermarkets compete to improve the quality and utility in order to achieve dissemination of information easily and quickly which is accurate and effective. This situation encourages the development of techniques that automatically find the relationship between item in the database. This study aims to analyzing and knowing association rules formed by using apriori algorithm. Market basket analysis’s steps are doing descriptive analysis, grouping the data transactions, applying apriori algorithm on the data, calculating the value of support and calculating the value of confidence. With the value of the minimum support 10% and minimum value of confidence 40%, the results obtained are one rule of association on the first day, four rules of association on the second day, one rule of association on the third day, four rules of association on the fourth day, six rules of association on the fifth day, nine rules of association on the sixth day, and four rules of association on the seventh day.

Perbandingan Metode Bootstrap Dan Jackknife Resampling Dalam Menentukan Nilai Estimasi Dan Interval Konfidensi Parameter Regresi Dessy Ariani; Yuki Novia Nasution; Desi Yuniarti
EKSPONENSIAL Vol 8 No 1 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (549.691 KB)

Regression analysis is a study that describes and evaluates the relationship between an independent variable and the dependent variable for the purpose of estimating or predicting the value of the dependent variable based on the value of the independent variables. Resampling is used when samples obtained for analyzing is less. In this study, Bootstrap method and Jackknife method are using. Both methods are used to find the value of regression parameter estimates and confidence intervals of regression parameter values which applied to the data position of Public Deposits in four groups of banks : Persero Banks, Government Banks, National Private Banks and Foreign Banks to knowing the best resampling methods to find the value of regression parameter estimates and confidence intervals of regression parameter values. There are three independent variables which are used in this study, namely investments loans, working capital loans and consumer loans. From the research results, it is obtained that the Jackknife method is the most appropriate method because it has smaller standard error values so Jackknife methods have a narrow range confidence intervals.

Klasifikasi Persediaan Barang Menggunakan Analisis Always Better Control (ABC) dan Prediksi Permintaan dengan Metode Monte Carlo Ricca Noviani; Yuki Novia Nasution; Nanda Arista Rizki
EKSPONENSIAL Vol 8 No 2 (2017)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (711.284 KB)

ABC analysis is a method of inventory control to manage a small number of item but has a high utilization. Inventories are categorized into three classes A, B, and C. The objective of the research is to manage the inventories using ABC analysis, EOQ, ROP, and to provide an overview of the next-year demand of the drug items using Monte Carlo method. ABC analysis results show that out of 79 drug items, class A consists of 19 drug items with usage value 69,11%, class B consists of 19 drug items with usage value 20,29%, and class C consists of 41 drug items with usage value 10,60%. Based on economic order quantity method, minimum ordering quantity of drug are two items and maximum ordering quantity of drug are 96 items.Based on reorder point method, the minimum quantity of drug for reordering is zero item and the maximum quantity of drug for reordering are seven items. Monte Carlo method results show that Fludane Plus 60 ml has the minimum demand on January - Desember 2017 which is only one bottle a month and Actifed Cough Merah 60 ml has the maximum demand which is 78 - 81 bottles a month. Lapisiv 100 ml, Kamulvit B12 Sirup 120 ml, Fludane Plus 60 ml and Miconazole 2% has the highest accuration with the percentage of error 0% and Ikadryl DMP Sirup 100 mlhas the lowest accuration with the percentage of error 0,22%.

Title

Found 16 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 16 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 16 Documents
Search