Claim Missing Document
Check
Articles

Penerapan Metode Fuzzy Subtractive Clustering Nur Azizah; Desi Yuniarti; Rito Goejantoro
EKSPONENSIAL Vol 9 No 2 (2018)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (769.335 KB)

Abstract

Cluster analysis is a statistical analysis to classify the objects to be some clusters based on checked variables and similarity of character between the objects. Quality of human living or society has been influenced by many things. In reality, population density is very influential to the quality of human living because high population density will cause many problems that impact on deterioration of quality of human living. Fuzzy Subtractive Cluster (FSC) methods using the data as a candidate of cluster center, so that duty of computation is hanging on the number of data and is not hang at dimension of data. This study aims is to determine the results of FSC at clustering the district in East Borneo based on wide of the district and total of population in 2015. The result shows there is 8 until 24 districts which have high population density. From validity of cluster, it isfounded that the best result for clustering the district in East Borneo based on wide of the district and sum of citizen in 2015 is 2 clusters, there are narrow district with many citizen and wide district with few citizen.
Perbandingan Klasifikasi Metode Naive Bayes dan Metode Decision Tree Algoritma (J48) pada Pasien Penderita Penyakit Stroke di RSUD Abdul Wahab Sjahranie Samarinda Irene Lishania; Rito Goejantoro; Yuki Novia Nasution
EKSPONENSIAL Vol 10 No 2 (2019)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (554.727 KB)

Abstract

Classification is a technique to form a model of the data that has not been classified, then the model can be used to classify new data. Naive Bayes is a classification using probability method based on the Bayes theorem with a strong assumption of independence. The decision tree algorithm (J48) is an implementation of the algorithm (C4.5) that produces decision trees. In this research, will be compared the results of classification accuracy with the naive Bayes method and the decision tree algorithm (J48) in stroke patients. That is, a person who has stroke will be classified by using the data of patients in Abdul Wahab Sjahranie Samarinda Hospital with 7 factors, namely age, gender, blood pressure, diabetes mellitus, dyslipidemia, uric acid levels and heart disease. The results showed that the decision tree algorithm (J48) method has the higher level of accuracy than the method naive Bayes for stroke classification.
Multi-Attribute Decision Making dengan Metode Fuzzy Technique for Order Preference by Similarity to Ideal Solution (FTOPSIS) Oktri Mayasari; Yuki Novia Nasution; Rito Goejantoro
EKSPONENSIAL Vol 9 No 1 (2018)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (410.686 KB)

Abstract

Fuzzy TOPSIS is a method that is used for identifying solution from one limited alternative set. The basic principle is that the chosen alternative must have the shortest distance from the positive ideal solution and the furthest distance from the negative ideal solution to determine relative proximity from an alternative with optimal solution. Fuzzy numbers in this method give effectiveness to determine the value of decision matrix. The purpose of this research is to find out the recommendation of investment in ADHI, PTPP, WIKA, and WSKT stocks by using fuzzy TOPSIS method. The alternatives that is used in this research are four stocks in the building construction sector on LQ45, from February to July 2017 namely Adhi Karya (Persero) Tbk. (ADHI), PP (Persero) Tbk. (PTPP), Wijaya Karya (Persero) Tbk. (WIKA), and Waskita Karya (Persero) Tbk. (WSKT) with the attributes that consist of nine financial ratios, namely Earnings Per Share (EPS), Book Value Per Share (BV), Debt to Assets Ratio (DAR), Debt to Equity Ratio (DER), Return on Assets (ROA), Return to Equity (ROE), Gross Profit Margin (GPM), Operating Profit Margin (OPM) and Net Profit Margin (NPM) on June 2016. The result of the research with fuzzy TOPSIS analysis generates preference value from stocks of ADHI amount 0,1711, stocks of PTPP amount 0,6169, stocks of WIKA amount 0,6310, and stocks of WSKT amount 0,7488. The result of preference value shows that stocks of WSKT with the highest preference value become the best recommendation option to invest rather than the stocks of ADHI, PTPP, or WIKA.
Analisis Pengendalian Kualitas Produksi Menggunakan Peta Kendali U dan Diagram Kontrol Decision On Belief (DOB). Nurul Rahmahani; Rito Goejantoro; Desi Yuniarti
EKSPONENSIAL Vol 10 No 1 (2019)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (407.197 KB)

Abstract

Statistical quality control is a problem solving technique used to monitor, control, analyze, manage, and improve products. There are two kinds of control charts, namely the attribute control chart and the variable control chart. The Decision On Belief (DOB) control chart is an attribute control chart based on Bayes's Theorem. In this study, to determine the comparison of control chart U and the DOB control chart the degree of control sensitivity in identifying out of control data on the production quality control data banner of Lineza digital printing in Samarinda. Based on the result of the research, it is found that quality control using U control chart and DOB control diagram has not been statistically controlled because there is still data out of control and in better sensitivity level in detecting out of control data is a DOB control chart because this diagram detects 65% while the U control chart is only 15%.
Analisis Positioning dengan Menggunakan Multidimensional Scaling Nonmetrik Devy Sintya Putri; Sri Wahyuningsih; Rito Goejantoro
EKSPONENSIAL Vol 9 No 1 (2018)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (674.431 KB)

Abstract

As the era progresses, more and more smartphone brands are present in the market in which it is difficult for consumers to decide which smart phone brands are good among others. The aims of this research are to know the position of five brands of the smart phone based on the consumer perception by using multidimensional scaling analysis (MDS) and also to know the superiority for each of these smartphone brands based also on the consumer perception focused on the product attribute and consumer perception about smartphone brands which they mostly prefered. So the result indicates that the coordinate points got based on the consumer perception by using MDS analysis are as follows; Asus is (10,494,2525), Oppo is (-4,154; 3,591), Samsung is (-4,216; (- 3,979)), Sony is (4,188 ; (- 3,985)), and Xiaomi is (-6,312; 1,848). Among the five smart phone brands above, Xiaomi has an advantage on the most affordable price attribute. Samsung has an advantage on the attributes of good screen display results, better known brands, more beautiful designs, complete features, ease for use, and large memory capacity. The smart phone brands of Asus, Oppo and Sony have the advantage on the results of a good camera and good processor performance. It is the fact that the most superior smart phone brands based on the consumer’s perception data are the brand of Oppo and Samsung smart phones.
Pemodelan Geographically Weighted Regression (GWR) Dengan Fungsi Pembobot Tricube Terhadap Angka Kematian Ibu (AKI) Di Kabupaten Kutai Kartanegara Tahun 2015 Muhammad Rahmad Fadli; Rito Goejantoro; Wasono Wasono
EKSPONENSIAL Vol 9 No 1 (2018)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (702.1 KB)

Abstract

Maternal Mortality in Kutai Kartanegara is a geographical problem that suspected affected by geographical factor which the global regression cannot model the relation well between the main problem and its independent variable. Therefore, Geographically Weighted Regression (GWR) is used to solve it. Spatial statistics is a method for analyzing data that has spatial correlation. GWR Model is the locally of global regression which considering the geographical or location as the weighted function for estimating the parameters of models. The tricube weighted function is used for the weighting. From this study, the models are different from location to others with also has the independent variables. For Samboja, Muara Jawa, Sanga-Sanga, Anggana, Muara Badak, Marang Kayu, and Tabang which are not affected by the indicators. Loa Janan, Loa Kulu, Muara Muntai, Kota Bangun, Tenggarong, Sebulu, Tenggarong Seberang, Muara Kaman, and Kenohan have the Maternal Mortality affected by Hospital Ratio per 1.000 Pregnant Mothers (x1). Muara Wis, Kenohan, dan Kembang Janggut have the Maternal Mortality affected by Childbirth with Medical Help (x2). Muara Muntai, Muara Wis, Kota Bangun, Sebulu, Tenggarong, Muara Kaman, Kenohan, and Kembang Janggut have the Maternal Mortality affected by Health Care of Childbed (x4).
Peramalan Kebutuhan Bahan Baku Plat Besi Menggunakan Metode Runtun Waktu Autoregressive Integrated Moving Average (ARIMA) dan Meminimumkan Biaya Total Persediaan dari Hasil Peramalan Mengunakan Metode Period Order Quantity (POQ) Mulyta Anggraini; Rito Goejantoro; Yuki Novia Nasution
EKSPONENSIAL Vol 10 No 1 (2019)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (479.075 KB)

Abstract

ARIMA method is used to predict future patterns of data that is expected to approach the actual data. In the case of inventory control, the company must have a good planning system for forecasting results to get maximum benefit. Period Order Quantity Method is used to solve inventory problem and minimize the total inventory cost. The research objective are to predict how many iron plates which CV. Isakutama needs from January 2017 to Desember 2017 with ARIMA method and to minimize the predicted total inventory cost using Period Order Quantity method. Based on the research, the forecasting results of the iron plates for 12 months are 24, 24, 25, 24, 25, 25, 25, 25, 25, 25, 25 and 25 units, so that the total inventory cost is Rp.1,177,264,000 by providing them once every 52 days.
Peramalan Regarima Pada Data Time Series Yudha Muhammad Faishol; Ika Purnamasari; Rito Goejantoro
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 (166.802 KB)

Abstract

RegArima method is a modelling technique that combines the ARIMA model with a regression model which uses a dummy variable called regressors or variable regression. The purposes of this study was to determine the calendar variation models and application of the model to predict plane ticket sales in January 2016 - December 2017. Based on the data analysis show that ticket sales have seasonal pattern, ie an increase in ticket sales when Idul Fitri. First determine the regressors which is only affected by one feast day is Eid. Then do the regression model, where the dependent variable (Y) is the volume of plane ticket sales and the independent variable (X) is regressors, so the regression model is Ŷt=1.029+1.335 X. The results of analysis show that all parameters had significant regression model and then do a fit test the model, the obtained residual normal distribution and ineligible white noise, which means that it still contained residual autocorrelation. ARIMA modeling is then performed on the data regression residuals. Results of analysis performed subsequent residual own stationary ARIMA model estimation and obtained ARIMA (0,0,1) with all parameters of the model was already significant and conformance test models had also found and that the residual qualified white noise and residual normal distribution. So the calendar variation model was obtained by the method RegARIMA: Yt = 1.029,5 + 1.337,3 Dt + 0,28712 at-1 + at. Based on the model of those variations could be predicted on plane ticket sales for January 2016-December 2017.
Perbandingan Pengelompokan K-Means dan K-Medoids Pada Data Potensi Kebakaran Hutan/Lahan Berdasarkan Persebaran Titik Panas Athifaturrofifah Athifaturrofifah; Rito Goejantoro; Desi Yuniarti
EKSPONENSIAL Vol 10 No 2 (2019)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (786.777 KB)

Abstract

The cases of forest/land fires in Indonesia seem endless, almost every year in the dry season similar problems always occur. Some areas in Indonesia often occur in forest fires and result in losses of up to trillions of rupiah. Various ways have been made to help the government in minimizing the potential for forest or land fires, one of them is by monitoring hot spots. In this study using data hot spots with parameters of latitude, longitude, brightness, fire radiation power and confidence by using the method of grouping K-Means and K-Medoids. The difference between these two methods is that the K-means method uses the mean as the center of the cluster, while K-Medoids uses representative objects (medoids) as the center of the cluster. This study aims to compare the results of the grouping of K-Means method with K-Medoids by using 42 data. The results of this study indicate that the K-Means method produces Silhouette Coefficient scores greater than K-Medoids. So that, K-Means can provide more accurate grouping results with a greater Silhouette Coefficient value.
Peramalan Dengan Metode Fuzzy Time Series Markov Chain Yenni Safitri; Sri Wahyuningsih; Rito Goejantoro
EKSPONENSIAL Vol 9 No 1 (2018)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (418.6 KB)

Abstract

Forecasting is an activity to predict what will happen in the future with certain methods. Fuzzy time series is a method known as artificial intelligence used to predict the problem which the actual data is formed in linguistic values using fuzzy principles as its basis. This study discusses the method of fuzzy time series developed by Ruey Chyn Tsaur to predict the closing price of the shares of PT. Radiant Utama Interinso Tbk April 2017. Markov Chain fuzzy time series method is used to analyze a time series data which is a combination of fuzzy time series model with Markov Chain. Forecasting of closing stock price based on data from January 2011 to March 2017 for April 2017 is Rp 224,29,00. Markov Chain's fuzzy time series method to forecast the closing stock prices data from January 2011 to March 2017 has a 3,48% of MAPE value or has a 96,52% of precision forecast. The results show that the Markov Chain fuzzy time series has an excellent level of accuracy for forecasting the closing stock prices.
Co-Authors Abidin, Ahmad Aliful Aditiya Risky Tizona Amanah Saeroni Andrea Tri Rian Dani Annabaa Aulia, Muzizah Ardyanti, Hesti Ariessela, Syeli Astuti, Putri Sri Athifaturrofifah Athifaturrofifah Cahyani, Era Tri Candra, Yossy Christyadi, Santo Dani, Andrea Tri Rian Darnah Darnah Andi Nohe Darnah, Darnah Desi Yuniarti Deviyana Nurmin Devy Sintya Putri Dewi Wulan Sari Dini Elizabeth Dwi Agoes Setiawan Dwi Husnul Mubiin Dwi Indra Yunistya Dyah Arumatica Novilla Etri Pujiati Fatmi’aturro’isah, Nurul Febriyanti, Nur Afifah Fidia Deny Tisna Amijaya Gerald Claudio Messakh Hairi Septiyanor Hidayatullah, Aji Syarif Ika Purnamasari Ika Purnamasari Ilham Adnan Kasoqi Irene Lishania Irfan Fadil Isgiarahmah, Afryda Juliartha, Made Angga Katianda, Kristin Rulin Khairun Nida Khoiril Anwar Lupinda, Indah Cahyani M. Fathurahman Mahmudi Mahmudi Martua Tri Januar Sinaga Meiliyani Siringoringo Memi Nor Hayati Memi Nor Hayati Memi Nor Hayati Memi Nor Hayati Mochammad Imron Awalludin Muhammad Rahmad Fadli Muhammad Rais Muhammad Yafi Mulyta Anggraini Murdani, Endah Mulia Ni Wayan Rica A Novalia, Viona Nur Annisa Fitri Nur Azizah Nurdayanti Nurdayanti Nurhasanah Nurhasanah Nurmin, Deviyana Nurul Rahmahani Oktri Mayasari Permana, Jordan Nata Primantoro, Sudhan Putra, Eko Prasatyo Putri, Nurlia Sucianti Rachman, Dezty Adhe Chajannah Rahmaulidyah, Fatihah Noor Rinaldi, Rival Satriya, Andi M Ade Sekar Nur Utami Septilasse, Rebeka Norcaline Sifriyani, Sifriyani Siringoringo, Meiliyani Siti Mahmuda Soraya, Raihana Sri Wahyuningsih Sri Wahyuningsih Sri Wahyuningsih Sri Wahyuningsih Sri Wahyuningsih Suerni, Widya - Surya Prangga Suyitno Suyitno Suyitno Suyitno Syafitri, Febriana Syaripuddin Syaripuddin Syaripuddin Syaripuddin Wasono Wasono Wasono, Wasono Widyawati Widyawati Yenni Safitri Yudha Muhammad Faishol Yuki Novia Nasution Yuki Novia Nasution, Yuki Novia Yuliasari, Pratiwi Dwi Yuniarti, Desi