Claim Missing Document
Check
Articles

Optimizing Multilayer Perceptron with Cost-Sensitive Learning for Addressing Class Imbalance in Credit Card Fraud Detection Wowon Priatna; Hindriyanto Dwi Purnomo; Ade Iriani; Irwan Sembiring; Theophilus Wellem
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 8 No 4 (2024): August 2024
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v8i4.5917

Abstract

The increasing use of credit cards in global financial transactions offers significant convenience for consumers and businesses. However, credit card fraud remains a major challenge due to its potential to cause substantial financial losses. Detecting credit card fraud is a top priority, but the primary challenge lies in class imbalance, where fraudulent transactions are significantly fewer than non-fraudulent ones. This imbalance often leads to machine learning algorithms overlooking fraudulent transactions, resulting in suboptimal performance. This study aims to enhance the performance of Multilayer Perceptron (MLP) in addressing class imbalance by employing cost-sensitive learning strategies. The research utilizes a credit card transaction dataset obtained from Kaggle, with additional validation using an e-commerce transaction dataset to strengthen the robustness of the findings. The dataset undergoes preprocessing with RUS and SMOTE techniques to balance the data before comparing the performance of baseline MLP models to those optimized with cost-sensitive learning. Evaluation metrics such as accuracy, recall, F1 score, and AUC indicate that the optimized MLP model significantly outperforms the baseline, achieving an AUC of 0.99 and a recall of 0.6. The model's superior performance is further validated through statistical tests, including Friedman and T-tests. These results underscore the practical implications of implementing cost-sensitive learning in MLPs, highlighting its potential to significantly enhance fraud detection accuracy and offer substantial benefits to financial institutions.
Algoritma K-Means Clustering Untuk Rekomendasi Pemberian Beasiswa Bagi Siswa Berprestasi Sagala, Febry Sandrian; Mugiarso, Mugiarso; Priatna, Wowon
Journal of Students‘ Research in Computer Science Vol. 2 No. 2 (2021): November 2021
Publisher : Program Studi Informatika Fakultas Ilmu Komputer Universitas Bhayangkara Jakarta Raya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31599/jdy9g441

Abstract

Scholarships are given to underprivileged students or outstanding students through a selection involving certain criteria. The criteria include the average value of report cards, parents' income, distance from home to school, number of dependents of parents, condition of the house, and status of the house. This study aims to assist the selection team in determining the award of scholarships so that they can provide appropriate and inappropriate recommendations, taking into account 6 criteria. The problem is that the existing scholarships are only given to students who do not have a father. The K-Means Clustering Algorithm can help Cluster students who are not eligible and eligible to get scholarship recommendations. The dataset used was 145 instances from the MAS scholarship selection committee. Attaqwa 02 Babylon. The data is calculated and tested using the K-Means Clustering algorithm. The results of the test were 32 people were recommended as eligible and 113 people were not eligible. The K-Means Clustering Algorithm can help the selection team to determine the scholarship award.
Penentuan Pola Frekuensi Menggunakan Algoritma Apriori pada Sistem Informasi Berbasis Web Herlawati , Herlawati; Handayanto , Rahmadya Trias; Rejeki , Sri; Priatna , Wowon; Atika , Prima Dina; Rofiah , Syahbaniar; Retnoningsih , Endang; Saputra , Faisal Adi; Pradana , Galih Apriansha
Journal of Students‘ Research in Computer Science Vol. 3 No. 2 (2022): November 2022
Publisher : Program Studi Informatika Fakultas Ilmu Komputer Universitas Bhayangkara Jakarta Raya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31599/nxkyrb30

Abstract

Currently, technology can affect services in any field, both in areas such as salon services and sales of clothing products. How to plan a marketing strategy using the web based on service transaction data and products that are most often chosen by customers. Therefore, an information system for determining frequency patterns is needed using the website-based Apriori Algorithm method. The results of research on salons based on the type of beauty treatment obtained for the type of treatment with a minimum confidence = 70%, the first confidence value is 63% if the customer chooses to wash (shampoo), the customer chooses scissors, the second the confidence value is 100%, if the customer chooses to blow then chooses also cut, and the third confidence value is 86% if the customer chooses creambath then the customer chooses to cut too. Meanwhile at the shop clothes determining the frequency pattern of types of clothes with a minimum value of confidence = 70% so that the results include if a customer buys a veil, the customer buys a robe with a confidence value of 71.43% and if a customer buys khimar then the customer will also buy a robe with a confidence value of 78, 57%. With these results salon and clothing store owners can determine marketing strategies by providing the right product and service recommendations to customers.
Implementasi Deep Learning Untuk Rekomendasi Aplikasi E-learning Yang Tepat Untuk Pembelajaran jarak jauh Priatna, Wowon; Purnomo , Rakhmat; Putra , Tri Dharma
Jurnal Kajian Ilmiah Vol. 21 No. 3 (2021): September 2021
Publisher : Lembaga Penelitian, Pengabdian Kepada Masyarakat dan Publikasi (LPPMP)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (554.294 KB) | DOI: 10.31599/jki.v21i3.521

Abstract

The purpose of this study is to recommend e-learning applications that are appropriate for use in online learning in college environments. The large number of e-learning platforms used by lecturers for online lecture activities results in students being forced to use several e-learning applications depending on the lecturer who teaches the courses taken, for the university also finally gives lecturers policies for distance learning reports each finished giving the material. In this study the data collection method began by taking data from the faculty to find out which e-learning applications were widely used by lecturers, then distributing questionnaires to students and lecturers who used the e-learning application to measure the e-leaning application with the e-learning criteria. Appropriate. The data is then processed into a dataset. The algorithm used in implementing deep learning is Artificial Neural Network (ANN). For the implementation of ANN, 27 variables were determined from the e-learning criteria and 1 target. In this ANN stage, prediction was used with classifications based on preparation, training, learning, evaluation and prediction using the python programming. The results obtained in this study that the Moodle application gets the highest score with an accuracy of 97% to be used as a recommendation for e-learning applications that are appropriate for universities to conduct online lectures.
Algoritma First in First Out (FIFO) Untuk Perancangan Aplikasi Pemesanan Kaos Sablon Widianto, Ilham Rizky; Priatna, Wowon; Lubis, Hendarman
Jurnal Kajian Ilmiah Vol. 23 No. 2 (2023): May 2023
Publisher : Lembaga Penelitian, Pengabdian Kepada Masyarakat dan Publikasi (LPPMP)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31599/tva3pd96

Abstract

The purpose of this study is to solve the problem of screen-printing T-shirt shops. For manual screen printing t-shirt shops, customers often have to visit the store in person or contact them via chat or phone, often encountering the following issues when ordering t-shirts: B. Irregular orders for those who have placed an order in advance or who have been waiting for a long time. One way to solve the queuing problem is the FIFO algorithm. FIFO algorithms are methods for organizing, processing, and manipulating basic data structures in computer systems. The FIFO algorithm phases in this study begin with the data preparation phase, the Gantt cart process, and finally his FIFO wait time. The result of the FIFO stage translates into creating applications using the Java programming language, Android Studio, and the FireBase database. The results of this study can be applied to his FIFO algorithm for customer queues in ordering T-shirts. A t-shirt ordering application was tested using the white box method by running the test case in four passes. All tests passed, so you can use the ordering application based on the FIFO algorithm.
Bayadome Geotours (BATOUR) Prototype for Geosite Management at Bayah Dome Geopark, Banten Munandar, Tb Ai; Sri Lestari, Tyastuti; Handayani, Dwipa; Noe’man,, Achmad; Fathurrazi, Ahmad; Priatna, Wowon; Karyaningsih, Dentik; Kapriadi, Engkap
International Journal of Information Technology and Computer Science Applications Vol. 2 No. 1 (2024): January - April 2024
Publisher : Jejaring Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58776/ijitcsa.v2i1.119

Abstract

The objective of this study is to create a technology-driven application prototype, named "Bayadome Geotours," as a cutting-edge solution to enhance geotourism governance and environmental conservation in the Bayah Dome Geopark, Banten. This research advances the utilisation of information and geospatial technology to improve visitor experiences and bolster local community involvement. It achieves this through an emphasis on needs analysis, prototype design, implementation, and testing. The Bayadome Geotours prototype is specifically engineered to offer a dynamic and engaging tourism encounter. Geospatial navigation capabilities enable users to digitally explore geosites, while an intuitive user interface assures accessibility for visitors with different levels of knowledge. This programme offers precise and comprehensive geological information, providing a novel method to enhance comprehension of the geological resources found in the Bayah Dome Geopark. Bayadome Geotours is a good example of the value of local community involvement in geotourism administration. This application serves as both a travel guide and a venue for the exchange of knowledge, local narratives, and cultural heritage. Engaging the public in sharing information fosters a stronger connection between tourists and the environment, resulting in a beneficial influence on the preservation of geosites and the overall management of destinations. Prototype testing conducted using a unit testing methodology demonstrates the successful execution of all system functionalities. The JEST tool's test results confirm that the Bayadome Geotours application is prepared for distribution to the general user base. Nevertheless, there are obstacles in the way of effectively managing and modernising the application, as well as achieving general acceptance, that must be addressed in order to guarantee the ongoing triumph of this prototype. However, Bayadome Geotours has created significant opportunities for advancing sustainable geotourism governance.
Analisis Pola Faktor Penyebab Balita Stunting Pada Dinas Kesehatan Kota Bekasi Menggunakan Algoritma FP-Growth Khoirunnisaa, Nabiilah; Priatna, wowon; -, Rasim; Warta, Joni
JURNAL TEKNOLOGI DAN ILMU KOMPUTER PRIMA (JUTIKOMP) Vol. 7 No. 1 (2024): Jurnal Teknologi dan Ilmu Komputer Prima (JUTIKOMP)
Publisher : Fakultas Teknologi dan Ilmu Komputer Universitas Prima Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34012/jutikomp.v7i1.4761

Abstract

Stunting is a common long-term nutritional problem among children under five, especially in developing countries, including Indonesia. Recently, the term stunting has become a popular topic of discussion while diverting attention from the problems of malnutrition and obesity. This study aims to determine the results of the formation of rules or association rules in determining the causal factors that most affect stunting in toddlers at the Bekasi City Health Office. The method that will be used in this research is the FP-Growth Algorithm. With the application of the FP-Growth Algorithm, accurate results can be obtained in determining the rules for determining the causal factors that most affect stunting in toddlers. The dataset has 4575 data and will be evaluated using the lift ratio. This research produces 26 rules, the best of which are Not Getting Exclusive Breastfeeding, Thinness, and Malnutrition, with a minimum support value of 0.5, a confidence value of 0.6, and a lift ratio of 1.17. Thus, applying the FP-Growth Algorithm in this study is effective because it achieves a lift ratio value of more than 1.
Social entrepreneurship management: PKK moms' poverty alleviation strategy Sudiantini, Dian; Priatna, Wowon; Meutia, Kardinah Indrianna
Jurnal Inovasi dan Pengembangan Hasil Pengabdian Masyarakat Vol. 1 No. 2 (2023): Jurnal Inovasi dan Pengembangan Hasil Pengabdian Masyarakat
Publisher : CV. BImbingan Belajar Assyfa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61650/jip-dimas.v1i2.234

Abstract

The PKK women who are a part of this partnership program (PKM) are from the village of Rancanumpang Gedebage in Bandung City, West Java. First off, there is a serious lack of enthusiasm among PKK mothers. The motive in question is motivation for company management and entrepreneurship; they are also motivated to become employees of other parties who offer employment prospects. Lectures, talks, demonstrations, and questions and answers are the approaches employed. It is hoped that by taking part in this service project, PKK women in Rancanumpang Gedebage, Bandung City, West Java, will help to reduce poverty by contributing to the improvement of the family economy and by being able to make use of the local community by the idea of the social entrepreneur management strategy.
Pemberdayaan Kendaraan Bermotor Dalam Mendukung Potensi Wisata Desa Sawarna Berbasis Aplikasi Batour dan Sitaman Bumi Priatna, Wowon; Andi Lawrence Hutahaean, Johanes; Enggar Putera, dkk, Diaz
VIDHEAS: Jurnal Nasional Abdimas Multidisiplin Vol. 1 No. 2 (2023): Desember 2023
Publisher : VINICHO MEDIA PUBLISINDO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61946/vidheas.v1i2.42

Abstract

Penelitian ini bertujuan untuk mengkaji pemberdayaan kendaraan bermotor sebagai sarana penunjang potensi wisata di Desa Sawarna. Fokus utama penelitian adalah implementasi aplikasi BATOUR dan SITAMAN BUMI sebagai alat pendukung untuk meningkatkan efisiensi dan aksesibilitas dalam mengelola potensi wisata desa. Metode penelitian yang digunakan melibatkan studi literatur, survei lapangan, dan analisis data untuk memahami dampak pemberdayaan kendaraan bermotor terhadap perkembangan pariwisata di Desa Sawarna. Hasil penelitian diharapkan dapat memberikan kontribusi pada pengembangan strategi pemberdayaan wisata desa berbasis teknologi aplikasi, sehingga dapat meningkatkan daya tarik dan kesejahteraan masyarakat setempat.
Pelatihan Jaringan Komputer Dan Manajemen Industri Wowon Priatna
Jurnal Abdimas Ekonomi dan Bisnis Vol. 3 No. 1 (2023): JAMEB (Jurnal Abdimas Ekonomi dan Bisnis)
Publisher : Fakultas Ekonomi Dan Bisnis Universitas Bhayangkara Jakarta Raya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

computer and Industrial Network Management Training is a non-profit program that improves the skills of trainees in computer network management and computer network development in the Industrial Revolution 4.0 era, with the aim of solving the problems faced by provincial and transmigration workers. Office Training Center. The most widely discussed problem is limited knowledge about computer networks and network implementation in the industrial world. The proposed solution is to offer participants support and training including: cabling engineering, TCP/IP concepts, device sharing, routing, Windows server installation and network management. The purpose of the training is to improve employee skills in mastering computer network languages. This is a mandatory skill that must be possessed by graduates of the Provincial Manpower and Transmigration Training Center. Training activities will be carried out for 2 (two) days, studying training programs and more interactive exercise-based learning. The results of this PKM show that online competency qualifications in the management of the education industry increased by 19.8%, an increase of 21.8%. Keyword : Industrial Network Development, Industrial Revolution 4.0, Training Program
Co-Authors -, Rasim Ade Iriani Agung Nugroho Agung Nugroho Agus Hidayat Aida Fitriyani, Aida Ajif Yunizar Pratama Yusuf Alexander, Allan D Alhillah, Yumaris Alfi Andi Lawrence Hutahaean, Johanes Andri Fajriya Andry Fadjriya Annisa Oktavianti Hermadi Asep R. Hamdani Asep Ramdhani M Asep Ramdhani Mahbub Atika , Prima Dina Dhea Putri Aprilyana Dwi Budi Srisulistiowati Dwipa Handayani Eka Nur A’ini Endang Retnoningsih Enggar Putera, dkk, Diaz Faisal Adi Saputra Fajar Mukharom Fathurrazi, Ahmad Febry Sandrian Sagala Fefbiansyah Hasibuan Galih Apriansha Pradana Hadi Kusmara Hendarman Lubis Herlawati Herlawati Hindriyanto Dwi Purnomo Ikhsan Romli Ilham Rizky Widianto Intan Safira Irwan Sembiring Ismaniah Ismaniah Joni Warta Joni Warta Joniwarta Joniwarta, Joniwarta Jumi Saroh Hidayat Kapriadi, Engkap Karyaningsih, Dentik Khoirunnisaa, Nabiilah Kustanto , Prio Lubis, Hendarman M. Fadhli Nursal Mayadi Mayadi Meutia, Kardinah Indrianna Mugiarso Mugiarso, Mugiarso Muhammad Khaerudin Noe’man,, Achmad Nurjeli Nurjeli Pradana , Galih Apriansha Prima Dina Atika Purnomo , Rakhmat Purnomo, Rakhmat Putra , Tri Dharma Rahmadya Trias Handayanto Rakhmat Purnomo Rasim Ratna Salkiawati Rejeki , Sri Retnoningsih , Endang Rinaldi Tunnisia Ritzkal, Ritzkal Rofiah , Syahbaniar Sagala, Febry Sandrian Saputra , Faisal Adi Silvi - Siti Setiawati Siti Setiawati Siti Setiawati Siti Setiawati, Andika Yusuf Hidayat Sri Lestari, Tyastuti Sri Rejeki Sudiantini, Dian Sulistiyo, Dwi Suryadi Syahbaniar Rofiah Tb Ai Munandar, Tb Ai Theopillus J. H. Wellem Tri Dharma Putra Tri Dharma Putra Tyastuti Sri Lestari Tyastuti Sri Lestari Tyastuti Sri Lestari Tyastuti Sri Lestari Widianto, Ilham Rizky Wiyanto Wiyanto