JournalTOCs

Indonesian Journal of Applied Statistics
Number of Followers: 0

This is an Open Access Journal

Open Access journal
ISSN (Online) 2621-086X
Published by Universitas Sebelas Maret

[27 journals]

Back Matter
- Authors: Hasih Pratiwi
  PubDate: 2022-05-31
  DOI: 10.13057/ijas.v5i1.61628
  Issue No: Vol. 5, No. 1 (2022)
Pemilihan Metode Predictive Analytics dengan Machine Learning untuk
Analisis dan Strategi Peningkatan Kualitas Kredit Perbankan
- Authors: Aznovri Kurniawan, Ahmad Rifa'i, Moch Abdillah Nafis, Nimas Sefrida, Harry Patria
  Pages: 1 - 11
  Abstract: As a factor that determines bank’s profitability, loan quality, that is categorized based on debtor’s collectability classification, always gets attention and become main analysis topic in banking industry. Through recent development of statistics and data science, especially in predictive analytics using machine learning techniques, more comprehensive analysis and prediction in loan quality can be conducted. This research is intended to give example on application of predictive analytics using machine learning technique for analysis and strategy recommendation in increasing bank’s loan quality improvement. In this research, some machine learning classification methods are compared to conduct predictive analytics in loan quality with big data size (big data analytics). Computation result of different methods are compared and summarized, resulted in recommendation on most appropriate method to achieve this research objective. This research concluded that for imbalanced big data size such as bank’s loan collectability, Tree Ensemble method, further development of Decision Tree method that is commonly used in machine learning, is one of appropriate methods to get satisfactory result in this research. Imbalanced data that can result in false positive may be overcame by oversampling Synthetic Minority Oversampling Technique (SMOTE). This research scope is limited to analysis and prediction of debtor’s collectability for the next several months, combined with analysis and strategy recommendations based on product type, gender, and debtor’s occupation. Further predictive analytics for the next several years by including external factors, such as economic growth, is not covered in this research and possible to be conducted. As machine learning application in Indonesian banking industry analysis is still in early phase, this research is expected to become one of reference in application of predictive analytics using machine learning in banking industry. Keywords: predictive analytics; machine learning; loan collectability; loan quality
  PubDate: 2022-05-30
  DOI: 10.13057/ijas.v5i1.55483
  Issue No: Vol. 5, No. 1 (2022)
Analisis Faktor yang Berpengaruh terhadap Waktu Survival Pasien Penyakit
Ginjal Kronis menggunakan Uji Asumsi Proportional Hazard
- Authors: Assyifa Lala Pratiwi Hamid, Sri Subanti, Yuliana Susanti
  Pages: 12 - 18
  Abstract: Chronic kidney disease is a disease whose risk of death is always increasing. This disease was ranked as the 13th leading cause of death in Indonesia in 2017. One of the successful managements of chronic kidney disease can be seen from the possibility of survival of patients with chronic kidney disease. To identify the probability of survival of an object, survival analysis is used. One method of survival analysis that can be used to determine the survival time of patients with chronic kidney disease is Cox regression. Cox regression must satisfy the proportional hazard assumption, where the ratio of the two hazard values must be constant with time. The graphical method, namely the log-log graph, can be used to test the proportional hazard assumption, but the results are only used as a provisional estimate. In this study, the goodness of fit test was used to test the assumptions by calculating the correlation between the Schoenfeld residuals and the survival time rank. In conclusion, the variables of hypertension and haemodialysis frequency meet the proportional hazard assumption.Keywords: chronic kidney disease; Cox regression; goodness of fit; log-log graph; proportional hazard assumption
  PubDate: 2022-05-30
  DOI: 10.13057/ijas.v5i1.48121
  Issue No: Vol. 5, No. 1 (2022)
Pemodelan Kasus Kronis Filariasis di Indonesia Tahun 2019 Menggunakan
Geographically Weighted Negative Binomial Regression (GWNBR)
- Authors: Sri Rahayu Yogyana Sinurat, Ernawati Pasaribu
  Pages: 19 - 30
  Abstract: Filariasis is a mosquito-borne disease caused by filarial worms. In Indonesia, filariasis is the third most common vector-borne and zoonotic disease in the community. Patients who in the chronic stage will fell pain due to swelling and infection in the limbs so that it can ruin the daily activities, reduce work productivity and cause economic losses for both sufferers and the country. In 2019, there were 28 filariasis endemic provinces and only 6 non-endemic provinces. This shows that the treatment of filariasis has not been fully successful. This study aims to determine the general description of chronic cases of filariasis, identify spatial heterogeneity and analyze factors that influence the number of chronic cases of filariasis using GWNBR. The modeling results five provinces groups based on significant variables. Variables that have a significant effect in all provinces are the ratio of health facilities of 100,000 population, the percentage of regions with PHBS policies and the average humidity. Meanwhile, the significant variables in several provinces are the percentage of slum households, the percentage of poor people and the average air temperature.Keywords: filariasis; overdispersion; spatial heterogeneity; negative binomial; GWNBR
  PubDate: 2022-05-30
  DOI: 10.13057/ijas.v5i1.59127
  Issue No: Vol. 5, No. 1 (2022)
Analisis Sentimen dari Aplikasi Shopee Indonesia Menggunakan Metode
Recurrent Neural Network
- Authors: Herni Utami
  Pages: 31 - 38
  Abstract: Sentiment analysis on unbalanced data will cause classification errors where the classification results tend to be in the majority class. Therefore, it is necessary to handle unbalanced data. In this study, a combination of synthetic minority oversampling technique (SMOTE) and Tomek link methods will be used to handle unbalanced data. In this study, we use the Recurrent Neural Network (RNN) method to analyze the sentiment of Shopee application users based on review data. Shopee Indonesia application review data shows that around 80% of Shopee application users have positive sentiments and 20% have negative sentiments, which means the data is not balance. In this study, preprocessing process with combination of synthetic minority oversampling technique (SMOTE) and Tomek link method used to handle the condition. The performance of the result is quite good, namely 80% accuracy, 84.1% precision, 92.5% sensitivity, 30% specificity, and 88.1% F1-score. It is better than performance of sentiment analysis that without preprocessing to handle imbalanced data.Keywords: sentiment analysis; imbalanced data; Tomek link; SMOTE; RNN
  PubDate: 2022-05-30
  DOI: 10.13057/ijas.v5i1.56825
  Issue No: Vol. 5, No. 1 (2022)
Implementation of Transfer Learning for Covid-19 and Pneumonia Disease
Detection Through Chest X-Rays Based on Web
- Authors: Nindya Eka Apsari, Sugiyanto Sugiyanto, Sri Sulistijowati Handajani
  Pages: 39 - 47
  Abstract: Coronavirus disease 2019, known as COVID-19, attacks the human respiratory system caused by severe acute respiratory syndrome coronavirus-2 (SARS-Cov-2). COVID-19 disease and pneumonia show similar symptoms such as fever, cough, even headache. Diagnosis of pneumonia can be tested through diagnostic tests, including blood tests, chest X-rays, and pulse oximetry, while the diagnosis of COVID-19 recommended by WHO is with swab test (RT-PCR). But in fact, the swab test method takes a relatively long time, for about one to seven days, for the result, and is not cheap. For that, there needs to be a development that can be one of the options in diagnosing COVID-19 and pneumonia at once, especially since both diseases have similar symptoms. One option that can be done is the diagnosis using a chest X-ray. This research aims to detect COVID-19 disease and pneumonia through chest X-rays using transfer learning to increase the accuracy of disease diagnosis with a more efficient time. The architecture used is EfficientNet B0 with variations in optimization parameters, learning rates, and epochs. EfficientNet B0 Adam optimization with a learning rate of 0.001 in the 6th epochs is a great model that we obtained. Furthermore, the evaluation of the model got accuracy, precision, recall, and f1-score of 92%. Then the model visualization is done using Grad-CAM. To implement the best model, web application development is done to make it easier to detect COVID-19 disease and pneumonia.Keywords: COVID-19; pneumonia; EfficientNet; transfer learning; web
  PubDate: 2022-05-31
  DOI: 10.13057/ijas.v5i1.59442
  Issue No: Vol. 5, No. 1 (2022)
Structural Equation Model (SEM) dalam Pemodelan Kemiskinan di Pulau
Sumatera
- Authors: Hasrat Ifolala Zebua, Geni Andalria Harefa
  Pages: 48 - 57
  Abstract: Poverty is a serious issue that must be addressed immediately by countries in the world, including Indonesia. The Indonesian government has implemented a variety of poverty reduction projects, such as providing education and health insurance. The rising poverty rate is due to the poor quality of education and health care. On Sumatra, there are 5,83 million poor people or 22,06 percent of the total number of poor people in Indonesia. This statistic appears to be quite large, and the government should be concerned about it. Factors causing poverty such as education and health are latent variables that cannot be measured directly. The suitable statistical method used is Structural Equation Model (SEM). In SEM analysis, there are three types of model fit tests: measurement model fit with Confirmatory Factor Analysis (CFA), overall model fit, and structural model fit. The results indicated that the model was fit or suitable for the model's tests. From the SEM model that was formed, it was found that health had a negative and significant effect on poverty and education did not have a significant effect on poverty and 77 percent of the variation in poverty could be explained by the SEM model that was formed.Keywords: poverty; education; health; SEM; CFA
  PubDate: 2022-05-31
  DOI: 10.13057/ijas.v5i1.50493
  Issue No: Vol. 5, No. 1 (2022)
Perbandingan K-Nearest Neighbor dan Random Forest dengan Seleksi Fitur
Information Gain untuk Klasifikasi Lama Studi Mahasiswa
- Authors: Isran K Hasan, Resmawan Resmawan, Jefriyanto Ibrahim
  Pages: 58 - 66
  Abstract: Accreditation is a quality and feasibility assessment form in carrying out higher education. One of the factors that affect accreditation is the length of student study. In this study, the length of student study is classified by using the best attributes resulting from selecting information gain features. In optimizing the classification algorithm, we process the data by converting the original data into data that is ready to be mined. The next step is dividing the data into training and testing data so that the classification algorithm can be applied. This study gives the best four attributes, with K-nearest neighbor (K-NN) classification of 86.67% and random forest classification of 100%.Keywords: length of study; information gain; K-nearest neighbor; random forest
  PubDate: 2022-05-31
  DOI: 10.13057/ijas.v5i1.58056
  Issue No: Vol. 5, No. 1 (2022)
New Mathematical Properties of the Kumaraswamy Lindley Distribution
- Authors: Samy Abd Elmoez Mahommed, Salah M. Mohamed
  Pages: 67 - 77
  Abstract: The Kumaraswamy Lindley distribution is a generalized distribution that has many applications in various fields, including physics, engineering, and chemistry. This paper introduces new mathematical properties for Kumaraswamy Lindley distribution such as probability weighted moments, moments of residual life, mean of residual life, reversed residual life, cumulative hazard rate function, and mean deviation. Keywords: Kumaraswamy Lindley distribution; probability weighted moments; residual life; hazard rate; mean deviation
  PubDate: 2022-05-31
  DOI: 10.13057/ijas.v5i1.56206
  Issue No: Vol. 5, No. 1 (2022)