Article Per Year (5 Year)
p-Index From 2021 - 2026
1.503
P-Index
This Author published in this journals
All Journal JURNAL MEDIA INFORMATIKA BUDIDARMA Building of Informatics, Technology and Science INFOKUM TEPIAN Journal of Artificial Intelligence and Digital Business
Harpad, Bartolomius
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Economics, Econometrics & Finance Electrical & Electronics Engineering Engineering

Published : 8 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 8 Documents
Search

 1 
Prediksi Pengajuan Kredit Usaha Pada Koperasi Menggunakan Algoritma K-Nearest Neighbor Harpad, Bartolomius; Bustomi, Tommy
Building of Informatics, Technology and Science (BITS) Vol 4 No 1 (2022): June 2022
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v4i1.1626

Abstract

Cooperative activities have become the activities most needed by many people because they are related to money, cooperatives are places that provide loans to housewives and also workers in a certain area or environment, the lack of interest offered by this cooperative is considered very easy. and very helpful for many parties in facilitating financial affairs, especially in financial matters, because the convenience offered by the cooperative makes many interested people ask for the same thing resulting in vulnerability to fraud, the importance of making predictions on prospective new business loan applications can help reduce the worst risks from various risks that occur in the future, in this study the k-nearest neighbor algorithm will be used as a prediction algorithm for prospective business credit applications at cooperatives, the value obtained is the value of training data or data records of several previous customers so as to easier to know new data as test data in a study. The results found in this study for prospective business credit applications are "Not Eligible" seen to the closest value based on the smallest value with (closest distance) between one another as many as 3 distances, namely numbers 1, 2 and 3 where number 1 states "Not feasible ”, at the second closest distance stating “Eligible” and number 3 stating “Not Eligible”, the most results stated Not Eligible so that the decision value on new customers had to be rejected “Not eligible” to be accepted
Penerapan Data Mining Untuk Prediksi Perkiraan Hujan dengan Menggunakan Algoritma K-Nearest Neighbor Nursobah, Nursobah; Lailiyah, Siti; Harpad, Bartolomius; Fahmi, Muhammad
Building of Informatics, Technology and Science (BITS) Vol 4 No 3 (2022): December 2022
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v4i3.2564

Abstract

Rain is a condition where water droplets fall from clouds to the earth. In life, the presence of rain is highly anticipated, rain can help people who have a profession as farmers. Rain that occurs on a large scale will really provide obstacles for the community, in addition to hampering activities or activities especially those carried out on outdoor rain can also cause disaster for the community in the form of flooding. Estimating rain for the community is very important, knowing whether it will rain or not can make it easier for the community to anticipate the possibilities that may occur due to rain. However, in the process of delivering forecasts, there is often an uneven distribution of information and delays in conveying information to the public regarding whether or not rain will occur. The community should be able to independently predict whether or not rain will occur. Data processing should be done properly and correctly. Data mining is a way that can be done to assist in data processing. In this study, the settlement process will be carried out using the K-Nearest Neighbor (K-NN) algorithm. The results obtained show that the data testing decision is NO. In other words, data mining and the K-Nearest Neighbor algorithm can help the problem solving process
Pengelompokan Siswa Layak Penerima Beasiswa dengan Menerapkan Algoritma K-Means Clustering Data Mining Harpad, Bartolomius; Fahmi, Muhammad; Pahrudin, Pajar; Andrea, Reza
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 2 (2024): April 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i2.7394

Abstract

Scholarships are financial assistance given to individuals with the aim of assisting them in financing the education they are pursuing. To overcome the gap between upper middle economic communities and lower middle economic communities in obtaining quality education. The aim of this program is to provide opportunities for financially disadvantaged students to experience quality education. In the process that occurs in awarding scholarships, there should be a strong basic foundation in the process of determining and making decisions that occur. Where the process of providing scholarships carried out so far should not be given to students who truly deserve it. The problem or impact that occurs from this is that the scholarship program does not run in accordance with the program's objectives, namely helping in the economic gap for students. One way that can be used to resolve this problem is to review previous recipient data. Data mining is a process of re-excavating data. Excavation is carried out by reviewing all the information contained in the data. In this research, the cluster analysis method is used, which is a multivariate technique used to group objects based on their characteristics. Clustering is the process of grouping data. Where the grouping process carried out on data is a grouping that does not yet have a class target or is called unsupervised learning. The results obtained in the research show that there are 2 clusters from the application of the K-Means algorithm. In cluster 1 there are 6 students in it and in cluster 2 there are 4 students in it.
Pengolahan Data Penjualan Pakaian dengan Menerapkan Algoritma Apriori Data Mining Harpad, Bartolomius; Lailiyah, Siti; Yusika, Andi
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6478

Abstract

In its input, so far the use of sales transaction data has only been stored as an archive. In fact, the data can be utilized and processed into useful information to increase product sales or product innovation. In this case, sales data analysis needs to be carried out. With information about sales patterns, it can be seen what consumers buy most often. So from consumer purchasing patterns, decision making can also be done by the store related to the products to be sold. The data mining process in analyzing sales data, the Apriori algorithm can be utilized in the sales data process, namely by providing a relationship between sales transaction data. The data in question is sales data on clothes or pants that are ordered so that consumer purchasing patterns are obtained. Thus, the store can use the data to take suitable business actions. In this case, data can be used as a consideration to ensure the next sales strategy. The existence of information about sales patterns can find out what consumers buy most often. The products that are most often purchased by consumers are Hightwais Jeans Snow, Neda Tunik Full Buttons with 100% support for each product. by knowing the products that are most often purchased by consumers, the company can develop a strategy in determining the purchase of clothes and pants to maintain the availability of clothes and pants needed by consumers.
Eye Disease Classification Using Convolutional Neural Network (CNN) with Web-based MobileNetV2 Architecture Fahriawan, Muhammad; Pratiwi, Heny; Harpad, Bartolomius
INFOKUM Vol. 13 No. 03 (2025): Infokum
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58471/infokum.v13i03.2851

Abstract

The high prevalence of preventable eye diseases, such as cataracts, glaucoma, and diabetic retinopathy, emphasizes the importance of accessible and efficient diagnostic solutions. This research aims to develop a web-based eye disease classification system using a lightweight Convolutional Neural Network (CNN) architecture, MobileNetV2, to overcome computational limitations in real-time applications. CRISP-DM methodology is applied, including dataset preparation, transfer learning with MobileNetV2 and VGG16, model evaluation, and implementation using Flask. The dataset from Kaggle consisting of 4,217 eye fundus images with four classes (cataract, glaucoma, diabetic retinopathy, and normal) was divided into 80% training, 10% validation, and 10% testing. Data augmentation and normalization were performed to improve model generalization. The results showed MobileNetV2 achieved the highest accuracy (90.14%) with low computational requirements, outperforming VGG16 (89.66%) and CNN (86.78%). MobileNetV2 displays balanced precision (89-99%), recall (74-96%), and F1-score (81-99%) across all classes, especially excelling in diabetic retinopathy detection. Its efficiency on resource-constrained environments makes it ideal for web integration. The developed Flask-based application allows users to upload images for instant classification, bridging the healthcare access gap. This research proves the effectiveness of MobileNetV2 in combining high accuracy and computational efficiency, offering a scalable solution for early screening of eye diseases, especially in remote areas.
Pengembangan Game Edukasi Warisan Budaya Nusantara Menggunakan Kombinasi Forward Chaining dan Linier Congruent Method Pratama, Dion; Harpad, Bartolomius; Harianto, Kusno
RIGGS: Journal of Artificial Intelligence and Digital Business Vol. 4 No. 2 (2025): Mei - Juli
Publisher : Prodi Bisnis Digital Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/riggs.v4i2.693

Abstract

Studi ini bertujuan menciptakan sebuah permainan edukatif yang memperkenalkan warisan budaya Nusantara kepada generasi muda dengan cara yang interaktif dan menarik. Dengan memanfaatkan teknologi, permainan ini diharapkan menjadi sarana pembelajaran yang efisien untuk memperkenalkan kekayaan budaya Indonesia, seperti rumah tradisional, busana khas, tarian lokal, dan instrumen musik unik dari berbagai wilayah. Dengan cara ini, permainan ini tidak hanya menawarkan kesenangan, tetapi juga berperan dalam menjaga warisan budaya bangsa. Metode yang diterapkan dalam pengembangan permainan ini adalah model pengembangan multimedia dengan mengintegrasikan dua pendekatan utama, yaitu Forward Chaining dan Metode Kongruen Linear (LCM). Forward Chaining diterapkan sebagai mekanisme penalaran dalam pengambilan keputusan yang berbasis pada aturan untuk menentukan arah permainan dankonten edukasi yang disesuaikan dengan respons pengguna. Sementara itu, LCM digunakan untuk menghasilkan angka acak semu dalam sistem kuis, dengan tujuan meningkatkan variasi soal dan pilihan jawaban, sehingga menciptakan pengalaman bermain yang dinamis dan bukan monoton. Hasil penerapan menunjukkan bahwa permainan edukasi ini dapat meningkatkan ketertarikan belajar pengguna terhadap materi budaya lokal, berdasarkan hasil pengujian yang dilakukan terhadap sekelompok responden siswa sekolah dasar. Permainan ini juga menunjukkan kinerja yang bagus dari segi teknis, dengan tampilan antarmuka yang ramah pengguna dan mekanisme permainan yang mudah dimengerti. Sebagai kesimpulan, dapat diketahui bahwa perpaduan antara metode Forward Chaining dan Linear Congruent Method efektif dalam mendukung pengembangan permainan edukatif yang berlandaskan budaya Nusantara.
Determining the Country with the Best Economic Conditions 2025 using the MCDM Method Harpad, Bartolomius; Azahari, Azahari; Salmon, Salmon
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7346

Abstract

In the midst of increasingly complex global challenges in 2025, evaluating a country's economic condition is an important element in supporting strategic decision-making, whether at the government, corporate or individual level. The diversity of economic indicators such as Gross Domestic Product (GDP), inflation, unemployment, and human development index often makes it difficult to make an objective and comprehensive assessment. Reliance on a single indicator tends to produce a biased and unrepresentative picture. To address these issues, this research adopts a Multi-Criteria Decision Making (MCDM) approach that is able to consider various economic aspects simultaneously and systematically. The three MCDM methods used in this study are TOPSIS, VIKOR, and COCOSO. The analysis was conducted on 19 countries using four main indicators, namely GDP in billion USD, inflation rate, unemployment rate, and economic growth rate. Based on the results of data processing, the USA occupies the top position as the country with the best economic performance, followed by China. The three methods show consistency in ranking some countries, but there are also striking differences for some alternatives due to different approaches in normalisation and weighting. These findings emphasise the importance of choosing the right method in multicriteria evaluation. Therefore, a combined approach such as ensemble decision-making is recommended to strengthen the validity of the results. For further development, the use of additional indicators and the integration of artificial intelligence-based technology are suggested to improve accuracy and flexibility in analysing economic conditions between countries.
Development Intelligent Agent in Educational Game “Pesut Adventure – Borneo Animal Match-Up” with Shuffle Random Algorithm Khoirunnita, Aulia; Harpad, Bartolomius; Ikhsan, Nurul; Andrea, Reza; Beze, Husmul; Rudito, Rudito
TEPIAN Vol. 4 No. 4 (2023): December 2023
Publisher : Politeknik Pertanian Negeri Samarinda

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51967/tepian.v4i4.2891

Abstract

Research developing Edu-game ”Pesut Adventure – Borneo Animal Educational Game” is a research develop Match-Up type game. In this type game, player must find match 2 images of the Borneo animals in the same time, the player must remember the position of the image to be matched. The shuffling-random algorithm used to make images position always scrambled and player never get bored playing. AI technology (artificial intelligence) is also applied on this research. Using the Finite State Machine (FSM) model, the game agent was created in funny-animals form. It will mentoring the children to play this game like a teacher
Co-Authors Azahari, Azahari Bustomi, Tommy Fachry Abda El Rahman Fahriawan, Muhammad Harianto, Kusno Husmul Beze, Husmul Khoirunnita, Aulia Muhammad Fahmi Nursobah, Nursobah Nurul Ikhsan Pahrudin, Pajar Pratama, Dion Pratiwi, Heny Reza Andrea Rudito Rudito Salmon Salmon, Salmon Yusika, Andi