Article Per Year (5 Year)
p-Index From 2021 - 2026
1.642
P-Index
This Author published in this journals
All Journal Pixel : Jurnal Ilmiah Komputer Grafis Jurnal Informatika International Journal of Artificial Intelligence Research Jurnal Informatika MUST: Journal of Mathematics Education, Science and Technology Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI) Jurnal Abdimas Prakasa Dakara Literasi: Jurnal Pengabdian Masyarakat dan Inovasi
Agus Subekti
Universitas Nusa Mandiri
Humanities Computer Science & IT Education Electrical & Electronics Engineering Engineering Mathematics Social Sciences Other

Published : 9 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 9 Documents
Search

 1 
PREDIKSI TINGKAT PELANGGAN CHURN PADA PERUSAHAAN TELEKOMUNIKASI DENGAN ALGORITMA ADABOOST Iqbal Muhammad Latief; Agus Subekti; Windu Gata
Jurnal Informatika Vol 21, No 1 (2021): Jurnal Informatika
Publisher : IIB Darmajaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30873/ji.v21i1.2867

Abstract

With the rapid advancement of the telecommunications industry, and competition between telecommunications companies is increasing, companies need to predict their customers to determine the level of customer loyalty. One of them is by analyzing customer data by doing a Customer Churn Prediction. Predicting Customer Churn is an important business strategy for the company. To acquire new customers is much higher cost than retaining existing customers. The ease of operator switching is one of the serious challenges that the telecommunications industry must face. By predicting customer churn, companies can take immediate action to retain customers. To retain existing customers, the company must improve customer service, improve product quality, and must know in advance which customers have the possibility to leave the company. Prediction can be done by analyzing customer data using data mining techniques. In line with this, gathering information from the telecommunications business can help predict whether customer relationships will leave the company. The data used in this study are secondary data and amount to 7.403 data customers. The data has 21 variables. This study proposes to use the ensemble method namely adaboost, xgboost and random forest and compare them. Algorithm is validated through training data and testing data with a ratio of 80:20. From the results we got using python tools, it was found that the adaboost algorithm has an accuracy of 80%.Keywords—accuracy, adaboost, churn prediction, compare model, data mining.
Penyuluhan Literasi Media untuk Bijak di Media Sosial dan Pemanfaatan Media Digital Dwiza Riana; Agus Subekti; Hilman F. Pardede; Zico Pratama Putra; Faruq Aziz
Jurnal Abdimas Prakasa Dakara Vol. 2 No. 2 (2022): Literasi Media dan Promosi Kreatif dalam Kegiatan Kemasyarakatan
Publisher : LPPM STKIP Kusuma Negara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37640/japd.v2i2.1522

Abstract

Understanding the power of media must be promoted at all levels. Efforts to develop media literacy, both in the form of thoughts and in conducting outreach activities, need to be carried out and supported by various stakeholders. Especially in the current era of digital media, people are used to and easily access social media. There is also growing concern about the negative impact of social media use on young people. Therefore, it is necessary to teach the younger generation media skills to use social media. This is the basis for making joint activities aimed at educating the younger generation to be wise in using social media and being able to use digital media well. This activity took place on April 3, 2022 with a total of 15 participants. Based on the results of the activities carried out, the application of positive communication resulted in positive changes in the insights, knowledge, skills, values, and attitudes of adolescents, and this activity has important benefits for community activities. to successfully achieve the goals and benefit the community, especially the partners of the SIGMA Foundation.
ANALISIS PERFORMA DAN KECEPATAN KOMPUTASI ALGORITMA K-MEANS DAN K-MEDOIDS PADA TEXT CLUSTERING Karno Nur Cahyo; Agus Subekti; Muhammad Haris
Pixel :Jurnal Ilmiah Komputer Grafis Vol 15 No 2 (2022): Vol 15 No 2 (2022): Jurnal Ilmiah Komputer Grafis
Publisher : STEKOM PRESS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/pixel.v15i2.931

Abstract

The large number of theses will certainly make it difficult to find categories on thesis topics that have been written by students at a university. One of the uses of the Text Mining method is being able to group thesis objects into the number of clusters formed by the clustering algorithm. This study aims to compare 2 clustering algorithms, namely the K-Means and K-Medoids algorithms to obtain an accurate evaluation of the performance and computational time in the case of thesis clustering, so that relevant topics can be grouped and have better clustering accuracy. The evaluation parameter used is the Davies Bouldin Index (DBI) which is one of the testing techniques on clustering results, with the distribution of training data and testing data using cross validation using a repetition parameter of 10 folds iteration. From the results of the study with the Term Weighting condition used is Term Occurrences and using the N-Grams value is 2, it can be concluded that the K-Means algorithm has a better DBI value of -0.426. Meanwhile, the range of DBI values owned by K-Medoids with the same conditions has a DBI value of -1,631. However, from the visualization results using t-SNE with the same supporting parameters, there are options that can be used, namely the number of clusters is 6, and the DBI value is -1.110. For testing the computational time in the clustering process of 50 thesis documents, the K-Means algorithm has an average time of 2.5 seconds while the K-Medoids algorithm has an average time of 261.5 seconds. The computer specifications used are Asus ZenBook UX425EA.312 with the processor used is 11th Gen Intel® Core™ i5-1135G7 @ 2.40GHz @ 2.40GHz, the graphics card is Intel® Iris® Xe Graphics, the RAM used is 8GB, with storage of 512GB SSD.
Prediksi Kinerja Siswa Pada E-Learning Moodle Platform Menggunakan Algoritma Adaptive Boosting Jordy Lasmana Putra; Agus Subekti
Jurnal Informatika Vol 10, No 1 (2023): April 2023
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v10i1.15525

Abstract

Pandemi Covid-19 yang sudah berlangsung sejak awal tahun 2020 memberikan dampak besar di berbagai sektor, salah satunya di sektor pendidikan, dimana awalnya pendidikan dilakukan secara tatap muka, karena pandemi mengharuskan proses belajar mengajar dilakukan secara dalam jaringan (daring) Teknologi informasi berkembang sangat pesat dan mempengaruhi berbagai bidang, salah satunya bidang pendidikan, yang dimana pembelajaran secara daring sudah menjadi hal yang biasa untuk era saat sekarang ini, salah satu Learning Management System atau yang sering disingkat LMS yang sering digunakan adalah E-Learning menggunakan platform moodle, ditambah untuk saat pandemic covid-19 proses pembelajaran diarahkan ke sistem daring, sehingga penggunaan E-Learning menjadi meningkat. Melihat hal tersebut penulis bermaksud untuk melakukan penelitian untuk melakukan prediksi terhadap kinerja siswa dalam mengikuti perkuliahan e-learning yang menggunakan moodle platform, penelitian ini melihat dari sisi log activity siswa di moodle platform lalu log tersebut di transformasi agar dapat dilakukan proses klasifikasi oleh algoritma machine learning. Pada penelitian ini penulis melakukan klasifikasi menggunakan algoritma Adaptive Boosting dengan Base Learner C4.5 dengan teknik pra pemrosesan data Resample untuk Imbalance data. Hasil dari penelitian ini didapatkan hasil performansi yang baik, dengan nilai Akurasi 95%, ROC 0.97, dan Kappa 0.90. sehingga penelitian ini dapat menjadi model untuk memprediksi kinerja siswa dengan melihat log aktivitasnya menggunakan platform moodle. The Covid-19 pandemic, which has been going on since the beginning of 2020, has had a major impact in various sectors, one of which is in the education sector, where initially education was carried out face-to-face, because the pandemic requires the teaching and learning process to be carried out online Information technology is developing very rapidly and affecting various fields, one of which is the field of education, where online learning has become commonplace for today's era,  one of the Learning Management Systems or often abbreviated as LMS that is often used is E-Learning using the moodle platform, plus during the Covid-19 pandemic the learning process is directed to an online system, so that the use of E-Learning becomes increasing. Seeing this, the author intends to conduct research to predict student performance in participating in e-learning lectures using the moodle platform, this study looks at the student activity log on the moodle platform and then the log is transformed so that the classification process can be carried out by machine learning algorithms. In this study, the authors classified using the Adaptive Boosting algorithm with Base Learner C4.5 with the Resample data preprocessing technique for data imbalance. The results of this study obtained good performance results, with an Accuracy value of 95%, ROC 0.97, and Kappa 0.90. So this study can be a model to predict student performance by looking at their activity logs using the Moodle platform.
Mobilenet-based Transfer Learning for Detection of Eucalyptus Pellita Diseases Deviana Sely Wita; Agus Subekti
Jurnal Nasional Pendidikan Teknik Informatika : JANAPATI Vol. 12 No. 1 (2023)
Publisher : Prodi Pendidikan Teknik Informatika Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/janapati.v12i1.53220

Abstract

Currently, the pulp industry in Indonesia is ranked eighth in the world and the paper industry is ranked sixth in the world. One of the advantages in supporting the industry is that Indonesia has a large Industrial Plantation Forest (HTI) where the plants for pulp and paper raw materials originate. Eucalyptus pellita species belonging to the Myrtaceae family is one of the priority species for Industrial Plantation Forests (HTI) because of its adaptability and its wood can be used as raw material for pulp. Industrial Plantation Forests of this type can be found mainly in Kalimantan and Sumatra. This species shows good growth in stem shape, growth speed and good wood quality and has high germination and has a shorter cutting cycle of about 7-8 years so that it is quickly harvested. Prevention and treatment of leaf disease is one of the main processes of planting. Early diagnosis and accurate recognition of Eucalyptus Pellita disease can control the spread of the disease and reduce production costs and treatment costs. Disease detection on Eucalyptus pellita leaves can be done automatically faster by utilizing digital image processing and artificial intelligence. In this study, we propose a detection method with Deep Learning architecture. Our proposed method is based on pre-trained transfer learning using MobileNet. Image datasets from PT. Surya Hutani Jaya's land in East Kalimantan were used to train the model. The dataset is divided into three classes where 1 class is healthy leaves and 2 classes are sick leaves, namely Xanthomonas Bacteria and Cylindrocladium Fungi. With a dataset ratio of 70: 20: 10 the number of training datasets is 2370, validation is 591, and Testing is 177. Hyperparameter scenarios were carried out on the MobileNet model to optimize performance on the Eucalyptus Pellita leaf dataset. The experimental results show a fairly good accuracy, reaching 98%.
Pelatihan Aplikasi E-Learning “MojadiPro” Berbasis Android untuk Siswa SMK IT Raflesia Depok Agus Subekti; Zico Pratama Putra; Dwiza Riana; Hilman Ferdinandus Pardede
Jurnal Pengabdian Masyarakat dan Inovasi Vol 3 No 1 (2023)
Publisher : Pengelola Jurnal Politeknik Negeri Ketapang Jl. Rangga Sentap, Dalong Sukaharja, Ketapang 78813. Telp. (0534) 3030686 Kalimantan Barat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58466/jurnalpengabdianmasyarakatdaninovasi.v3i1.1194

Abstract

SMK IT Raflesia Depok is one of the high schools in the Depok City area which currently has a website that is used as a medium of information and communication, but does not yet provide teaching material services online. Apart from getting subject matter from school, students are usually still looking for additional material by attending private classes. With the development of technology that has developed in the field of education, access to subject matter can be done online, or what is commonly called e-learning. In order to carry out the Tri Dharma of Higher Education, Nusa Mandiri University held an android-based "MojadiPro" e-learning application training for students of SMK IT Raflesia Depok which aims to provide information regarding the existence of the MojadiPro e-learning application, which later can be used by students as a companion application learning that can be accessed via a smartphone. The training was held on January 14 2023 at UNM Margonda Campus with 15 participants. The research results can increase participants' understanding of the development of e-learning from time to time and the technology used in the development of e-learning platforms
Date Palm Identification using DenseNet-201 Transfer Learning Method Kusnadi Bin Raman; Agus Subekti
Jurnal Nasional Pendidikan Teknik Informatika : JANAPATI Vol. 12 No. 3 (2023)
Publisher : Prodi Pendidikan Teknik Informatika Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/janapati.v12i3.65932

Abstract

There are more than 400 types of dates in the world that are similar in size, shape, colour, fruit texture, taste and maturity, making it difficult for people to memorise them. Identification with artificial intelligence can make labelling dates easier. This research proposes the DenseNet-201 transfer learning method with freeze all the pre-trained layers, re-train all the pre-trained layers, and hyperparameter models for date variety identification. The date dataset was collected from the market with a total of 3,300 images of 11 types of dates, including Ajwa, Bam, Golden, Khalas, Khenaizi, Lulu, Mabroum, Medjool, Safawi, Sukari and Tunisian. The purpose of the research is to identify, analyse the test images and compare and recommend the best performance model to identify the type of dates. The experimental results have resulted in the recommendation that the DenseNet-201 method with the hyperparameter model shows the best performance with an accuracy value of 99.39%.
Visual Historical Data-Based Traffic Movement and Density Pattern Extraction For Adaptive Pattern Detection Base On Vehicle Type Filda Angellia; Nita Merlina; Agus Subekti; Rahmadya Trias Handayanto
International Journal of Artificial Intelligence Research Vol 10, No 1 (2026)
Publisher : Universitas Dharma Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29099/ijair.v8i2.1715

Abstract

Traffic congestion in urban areas has become a crucial issue, impacting time efficiency, energy consumption, and quality of life. One of the main causes of difficulties in traffic management is the lack of optimal predictive systems capable of detecting and adaptively responding to vehicle movement patterns. This study proposes a historical digital image-based approach to extract traffic movement patterns and density based on vehicle type and dimensions. The developed model utilizes historical traffic video footage from CCTV systems as a visual data source, which is then processed using the YOLOv5 algorithm to detect the number, size, and type of vehicles. After the detection process, vehicle information is converted into a sequential format that reflects vehicle movement in the temporal dimension. This data is then analyzed using a Long Short-Term Memory (LSTM) model to generate traffic density prediction patterns. This study also compares the performance of LSTM with other algorithms such as Random Forest and XGBoost in terms of prediction accuracy. Model evaluation is conducted using MSE and RMSE metrics to measure accuracy against actual data.The research results show that the integration of dimension-based vehicle detection with a visual historical data-driven prediction approach can improve the accuracy and flexibility of modeling future traffic conditions. This approach significantly contributes to the development of intelligent transportation systems that can adapt to dynamic environmental conditions and traffic patterns
Pendekatan Ensemble dan Voting untuk Prediksi Cacat pada Berbagai Proyek Perangkat Lunak Kirso Kirso; Agus Subekti
MUST: Journal of Mathematics Education, Science and Technology Vol 10 No 1 (2025): JULY
Publisher : Universitas Muhammadiyah Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30651/must.v10i1.26548

Abstract

This study conducted experiments using ensemble methods, hyperparameter tuning, and voting to improve software defect prediction across multiple projects using the Kamei dataset. Five machine learning models LightGBM, XGBoost, Random Forest, Extra Trees, and Gradient Boosting were applied to six projects: Bugzilla, Columba, JDT, Mozilla, Platform, and Postgres. Overall, the models demonstrated good performance when tested on datasets of projects with similar characteristics or strong relationships, such as Mozilla, JDT, and Platform, achieving accuracy and F1 scores above 80%. This indicates that defect patterns learned from one project can be effectively applied to similar projects. However, the models’ performance dropped significantly when predicting defects in the Bugzilla project from other projects, indicating notable differences in defect patterns or feature incompatibility. Differences in data distribution across projects remain a major challenge in CPDP. Therefore, domain adaptation techniques or feature transformation methods are needed to reduce inter-project differences, enabling the models to better recognize defect patterns across projects. Despite some improvements, data differences and class imbalance still limit prediction performance. Future research should address these challenges.
Co-Authors Deviana Sely Wita Dwiza Riana Faruq Aziz Filda Angellia Hilman F. Pardede Hilman Ferdinandus Pardede Iqbal Muhammad Latief Jordy Lasmana Putra Karno Nur Cahyo Kirso Kirso Kusnadi Bin Raman Muhammad Haris Nita Merlina Rahmadya Trias Handayanto Windu Gata Zico Pratama Putra