Article Per Year (5 Year)
p-Index From 2020 - 2025
2.949
P-Index
This Author published in this journals
All Journal Jurnal Buana Informatika Biogenesis: Jurnal Ilmiah Biologi JOURNAL OF APPLIED INFORMATICS AND COMPUTING JURNAL ILMIAH INFORMATIKA Journal of Computer System and Informatics (JoSYC) Journal of Information System and Technology (JOINT) Jurnal Pustaka Data : Pusat Akses Kajian Database, Analisa Teknologi, dan Arsitektur Komputer Jurnal Ilmiah Teknologi dan Komputer (JITTER) J-SIGN (Journal Of Informatics, Information System, And Artificial Intelligence) JoDENS: Journal of Digital Ecosystem for Natural Sustainability JUTEKOM - Jurnal Teknologi dan Ilmu Komputer
Purnajaya, Akhmad Rezki
Universitas Universal
Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Immunology & microbiology Industrial & Manufacturing Engineering Mechanical Engineering

Published : 15 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 15 Documents
Search

 1   2 
Performance Comparison of Data Sampling Techniques to Handle Imbalanced Class on Prediction of Compound-Protein Interaction Akhmad Rezki Purnajaya; Wisnu Ananta Kusuma; Medria Kusuma Dewi Hardhienata
Biogenesis: Jurnal Ilmiah Biologi Vol 8 No 1 (2020)
Publisher : Department of Biology, Faculty of Sci and Tech, Universitas Islam Negeri Alauddin Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/bio.v8i1.12002

Abstract

The prediction of Compound-Protein Interactions (CPI) is an essential step in the drug-target analysis for developing new drugs as well as for drug repositioning. One challenging issue in this field is that commonly there are more numbers of non-interacting compound-protein pairs than interacting pairs. This problem causes bias, which may degrade the prediction of CPI. Besides, currently, there is not much research on CPI prediction that compares data sampling techniques to handle the class imbalance problem. To address this issue, we compare four data sampling techniques, namely Random Under-sampling (RUS), Combination of Over-Under-sampling (COUS), Synthetic Minority Over-sampling Technique (SMOTE), and Tomek Link (T-Link). The benchmark CPI data: Nuclear Receptor and G-Protein Coupled Receptor (GPCR) are used to test these techniques. Area Under Curve (AUC) applied to evaluate the CPI prediction performance of each technique. Results show that the AUC values for RUS, COUS, SMOTE, and T-Link are 0.75, 0.77, 0.85 and 0.79 respectively on Nuclear Receptor data and 0.70, 0.85, 0.91 and 0.72 respectively on GPCR data. These results indicate that SMOTE has the highest AUC values. Furthermore, we found that the SMOTE technique is more capable of handling class imbalance problems on CPI prediction compared to the remaining three other techniques.
Ant Colony Optimization for Prediction of Compound-Protein Interactions Akhmad Rezki Purnajaya
Journal of Applied Informatics and Computing Vol 3 No 2 (2019): Desember 2019
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (952.028 KB) | DOI: 10.30871/jaic.v3i2.1639

Abstract

The prediction of Compound-Protein Interactions (CPI) is an essential step in drug-target analysis for developing new drugs. Therefore, it needs a good incentive to develop a faster and more effective method to predicting the interaction between compound and protein. Predicting the unobserved link of CPI can be done with Ant Colony Optimization for Link Prediction (ACO_LP) algorithms. Each ant selects its path according to the pheromone value and the heuristic information in the link. The path passed by the ant is evaluated and the pheromone information on each link is updated according to the quality of the path. The pheromones on each link are used as the final value of similarity between nodes. The ACO_LP are tested on benchmark CPI data: Nuclear Receptor, G-Protein Coupled Receptor (GPCR), Ion Channel, and Enzyme. Result show that the accuracy values for Nuclear Receptor, GPCR, Ion Channel, and Enzyme dataset are 0.62, 0.62, 0.74, and 0.79 respectively. The results indicate that ACO_LP has good accuracy for prediction of CPI.
Perbandingan Performa Teknik Sampling Data untuk Klasifikasi Pasien Terinfeksi Covid-19 Menggunakan Rontgen Dada Akhmad Rezki Purnajaya; Fuad Dwi Hanggara
Journal of Applied Informatics and Computing Vol 5 No 1 (2021): July 2021
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v5i1.3010

Abstract

The COVID-19 virus became a virus that was deadly and shocked the world. One of the consequences caused by the COVID-19 virus is a respiratory infection. The solution put forward for this problem is with a prediction of the COVID-19 virus infection. This prediction was made based on the classification of chest X-ray data. One challenging issue in this field is the imbalance on the amount of data between infected chest X-rays and uninfected chest X-rays. The result of imbalanced data is data classification that ignores classes with fewer data. To overcome this problem, the data sampling technique becomes a mechanism to make the data balanced. For this reason, several data sampling techniques will be evaluated in this study. Data sampling techniques include Random Undersampling (RUS), Random Oversampling (ROS), Combination of Over-Undersampling (COUS), Synthetic Minority Over-sampling Technique (SMOTE), and Tomek Link (T-Link). This study also uses the Support Vector Machines (SVM) data classification, because it has high accuracy. Furthermore, the evaluation is carried out by selecting the highest accuracy and Area Under Curve (AUC). The best sampling technique found was SMOTE with an accuracy value of 99% and an AUC value of 99.32%. The SMOTE technique is the best data sampling technique for the classification of COVID-19 chest x-ray data.
PENGENALAN SUARA PADA KAMUS BANJAR-INDONESIA DAN INDONESIA-BANJAR MENGGUNAKAN STATISTIK INFERENSI Akhmad Rezki Purnajaya; Fatma Indriani; Mohammad Reza Faisal
JURNAL ILMIAH INFORMATIKA Vol 8 No 01 (2020): Jurnal Ilmiah Informatika (JIF)
Publisher : Teknik Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (401.634 KB) | DOI: 10.33884/jif.v8i01.1727

Abstract

Banjar language used in conversation and daily life around the area. So foreigners who come to the regions of South Kalimantan will have difficulty in communicating. Besides, most local residents in the backwoods of South Kalimantan can not use Indonesian language properly, they would be more convenient to use regional language to interact. For that reason we need an Android application can help users to find the translation of a word or phrase whenever and wherever. With the help of Google Voice Search, this application can also listen to the voice of the user to be converted into text and insert into the input translation. Speech recognition of Banjar language required a literacy training data by using the method of statistical inference to make results appropriated. Testing using method of Black Box Testing to measure the percentage of suitability of the results of translation, speech recognition for Indonesian language and speech recognition Banjar language using method of Statistical inference. So the results of translation accuracy 100% and accuracy of speech recognition Indonesian language and Banjar language by 97.85% and 82.74%.
Implementasi Text Mining untuk Mengetahui Opini Masyarakat Tentang Climate Change Akhmad Rezki Purnajaya; Vinxencius Lieputra; Vincent Tayanto; Jaden Gil Salim
Journal of Information System and Technology (JOINT) Vol 3 No 3 (2022): Journal of Information System and Technology (JOINT)
Publisher : Program Sarjana Sistem Informasi, Universitas Internasional Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37253/joint.v3i3.7337

Abstract

Climate change adalah sesuatu yang terjadi dalam kehidupan kita sehari-hari dan sangat berpengaruh dalam kehidupan kita. Dampaknya cukup besar, seperti perubahan curah hujan, perubahan panjang musim dan sebagainya. Untuk menghadapi climate change, masyarakat memberikan opininya melalui Twitter dengan harapan adanya perubahan baik yang akan terjadi. Untuk mengetahui kesimpulan opini masyarakat tentang climate change dilakukan text processing dengan menggunakan metode text mining. Text mining adalah penambangan data berupa teks yang kemudian melalui beberapa tahapan seperti preprocessing sampai clustering. Hasil text mining yang diperoleh yaitu mengelompokan kata-kata opini pengguna twitter menjadi lima kelompok yaitu kelompok kata kunci, kelompok hastag yang sering digunakan, kelompok dampak yang dirasakan masyarakat, kelompok dampak utama climate change, dan kelompok objek alam yang dirugikan karena climate change.
Analisa Sentimen Informasi Hoaks Pasca Pandemi Covid-19 dengan Text Mining Akhmad Rezki Purnajaya; Yonky Pernando
Journal of Computer System and Informatics (JoSYC) Vol 4 No 3 (2023): May 2023
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v4i3.3358

Abstract

In this era of globalization, the delivery of information can be conveyed by any information media, this can also cause the spread of hoax information. This is very detrimental for recipients of information, especially information that is very important to receive and information related to the Covid or Corona virus that has hit the world and Indonesia. Based on data from the Gugus Tugas Percepatan Penanganan Covid-19, in the post-pandemic era 227 hoax information were still found that were identified. Therefore, in this study the authors conducted hoax information research, especially regarding the Covid-19 virus in the post-pandemic era by using the Text Mining method to find out information patterns that often appear in the post-Covid-19 pandemic. This paper will also show some results regarding the association of several words with the main word in hoax information in the post-Covid-19 pandemic era in the form of Term of Frequency, Word Cloud, Fruchterman Reingold Layout, and Circle Layout. The results of this study indicate that in the post-Covid-19 pandemic era, hoax information is frequently associated with ‘vaccines’ and ‘covid’, with 113 and 111 occurrences respectively. This is due to public skepticism regarding the safety and effectiveness of the Covid-19 vaccine, leading to the dissemination of hoaxes aimed at discouraging vaccination. Other commonly mentioned words in post-pandemic hoax information include ‘omicron’, ‘vaccinated’, ‘variant’, ‘pfizer’, ‘mRNA’, ‘virus’, and ‘vaccination’, albeit with lower frequencies
Analisis Ancaman COVID-19 Varian XBB di Indonesia Pada Jejaring Media Sosial Twitter Menggunakan Text Mining Purnajaya, Akhmad Rezki; ., Very; Noverio, Oscar; Alvaro, Charlos
JITTER : Jurnal Ilmiah Teknologi dan Komputer Vol 4 No 1 (2023): JITTER, Vol.4, No.1, April 2023
Publisher : Program Studi Teknologi Informasi, Fakultas Teknik, Universitas Udayana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (272.599 KB) | DOI: 10.24843/JTRTI.2023.v04.i01.p02

Abstract

Semakin cepatnya mutasi penyakit Covid-19 menyebabkan munculnya subvarian baru yang dikenal varian XBB pertama kali di Afrika Selatan pada tanggal 24 November 2021. Subvarian ini memiliki kekhasan khusus daripada subvarian lain dalam kecepatan penyebarannya yang sangat cepat, tetapi sebagian besar gejala yang didampakkan masih skala ringan. Hal ini menyebabkan kepanikan kembali oleh masyarakat Indonesia yang telah kembali melakukan aktivitas outdoor dengan normal dengan munculnya subvarian ini. Oleh karena itu penelitian ini bertujuan untuk menganalisis ancaman subvarian COVID-19 bernama XBB di Indonesia yang paling banyak dibicarakan orang di media sosial Twitter. Penelitian ini menggunakan metode Text Mining. Model yang digunakan dalam penelitian ini bervariasi, seperti matriks, word cloud, dan hierarchical clustering. Hasilnya menunjukkan bahwa dua kota di Indonesia terancam varian baru XBB yaitu Kota Bogor, Jawa Barat dan Kota Batam, Kepulauan Riau. Ditambah hasil analisa menunjukkan pemerintah Indonesia memberikan respon cepat untuk mencegah penyebaran dari subvarian XBB ini.
Pendekatan dengan Oversampling dan Undersampling untuk Meningkatkan Akurasi Diagnostik Kanker Tiroid Purnajaya, Akhmad Rezki; Darmawan, Justin; Yamin, Valerian; Charles, Charles
Jurnal Pustaka Data (Pusat Akses Kajian Database, Analisa Teknologi, dan Arsitektur Komputer) Vol 4 No 1 (2024): Jurnal Pustaka Data (Pusat Akses Kajian Database, Analisa Teknologi, dan Arsitekt
Publisher : Pustaka Galeri Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55382/jurnalpustakadata.v4i1.667

Abstract

Kanker tiroid, yang menjadi perhatian global dan sering kali tidak memiliki gejala awal, sehingga memerlukan deteksi yang tepat. Penelitian ini menggunakan Support Vector Machine (SVM) untuk identifikasi subtype kanker tiroid, yang bertujuan untuk meningkatkan akurasi, sensitivitas, dan spesifisitas. Dengan memanfaatkan data klinis, penelitian ini menggabungkan pemrosesan awal data dan memanfaatkan Random Over-Sampling (ROS) dan Random Under-Sampling (RUS) untuk mengatasi ketidakseimbangan kelas. Hasilnya menunjukkan kinerja klasifikasi yang tinggi, dengan data sampel menunjukkan sensitivitas yang unggul. Penerapan SVM yang sukses, bersama dengan ROS dan RUS, menjanjikan akurasi diagnostik yang lebih baik dan hasil pasien yang lebih baik.
Mango and Banana Ripeness Detection based on Lightweight YOLOv8 Saragih, Raymond Erz; Purnajaya, Akhmad Rezki; Syafrinal, Ilwan; Pernando, Yonky; Yodi
Jurnal Buana Informatika Vol. 15 No. 2 (2024): Jurnal Buana Informatika, Volume 15, Nomor 02, Oktober 2024
Publisher : Universitas Atma Jaya Yogyakarta

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

Abstract

Fruits like bananas and mangoes are harvested after reaching a specific ripeness stage. Traditionally, farmers rely on manual inspection to determine ripeness, a process that can be tedious, time-consuming, expensive, and subjective. This work proposes an automatic bananas and mangoes ripeness detector utilizing computer vision technology. The detected bananas and mangoes fall into two classes: ripe and unripe. The state-of-the-art YOLOv8 architecture serves as the core of the detector. Three YOLOv8 variants, YOLOv8n, YOLOv8s, and YOLOv8m, were investigated for their performance. Results show that YOLOv8s achieved the highest overall performance, 0.9991 recall, and a mean Average Precision (mAP) of 0.8897. While YOLOv8m achieved the highest precision of 0.9995, YOLOv8n is the most miniature model, making it suitable for deployment on devices with limited resources.
Perbandingan Performa Arsitektur Machine Learning untuk Deteksi Dini Depresi Berbasis Natural Language Processing dalam Bahasa Indonesia Pangestu, Amora Antonio; Purnajaya, Akhmad Rezki
J-SIGN (Journal of Informatics, Information System, and Artificial Intelligence) Vol 3, No 2 (2025): November
Publisher : Department of Informatics, Universitas Syiah Kuala

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24815/j-sign.v3i2.49873

Abstract

Depresi telah menjadi krisis kesehatan mental global yang mendesak. Pemanfaatan Natural Language Processing (NLP) pada analisis teks digital menawarkan potensi besar untuk deteksi dini depresi secara non-intrusif. Penelitian ini menyajikan analisis komparatif dari tiga arsitektur machine learning, yaitu Naive Bayes, Long Short-Term Memory (LSTM), dan Convolutional Neural Network (CNN) untuk mengklasifikasikan teks berbahasa Indonesia. Metodologi penelitian dimulai dengan akuisisi dan pra-pemrosesan (pembersihan, case folding, tokenisasi, stopword removal) dataset 10.801 teks yang teranotasi psikolog. Model dilatih pada 75% data dan dievaluasi pada 25% data uji menggunakan metrik akurasi, presisi, recall, dan F1-score. Hasil menunjukkan bahwa arsitektur CNN mencapai performa tertinggi dengan F1-score seragam sebesar 94%. Kinerja ini sedikit melampaui model LSTM (93%) dan secara signifikan mengungguli Naive Bayes (86%). Keterbatasan penelitian ini mencakup fokus pada klasifikasi biner dan belum digunakannya arsitektur Transformer. Temuan ini memberikan landasan penting untuk pengembangan deteksi kesehatan mental di Indonesia yang lebih akurat, adaptif, efektif, dan relevan terhadap konteks budaya lokal.
Co-Authors ., Very Alvaro, Charlos Charles Charles Darmawan, Justin Eva Tesvara Fatma Indriani Fery Gunawan Fuad Dwi Hanggara Gunawan, Fery Ilwan Syafrinal Jaden Gil Salim Jaya, Mega Joni Irwansyah Kaharuddin Limuel, Patrick Medria Kusuma Dewi Hardhienata Milson Nestelrody Muhammad Reza Faisal, Muhammad Reza Noverio, Oscar Pangestu, Amora Antonio Patrick Limuel Pernando, Yonky Raymond Erz Saragih Rika Jelita Roza, Yuni Saragih, Raymond Erz Steffany Tayanto, Vincent Vincent Tayanto Vinxencius Lieputra Williams Williams Wisnu Ananta Kusuma Yamin, Valerian Yodi Yuni Roza