Huzain Azis

Full Author List

Baharuddin, Mus Mulyadi; Azis, Huzain; Hasanuddin, Tasrif; Unviersitas Muslim Indonesia, Indonesia

Abstract

Nowadays, the industry makes various types of goods that have glass-based materials, float car window panes, non-float building windows, lamps, jars, and tableware. These glasses have the same production material, the difference between one and the other is the composition of the production material. K-Nearest Neighbor (KNN) algorithm which is one of the classification methods in data mining and also a supervised learning algorithm in machine learning is a method for classifying objects based on learning data that is the closest distance to the object.. This study discusses the performance measurement (accuracy, precision, recall and f-measure) of the KNN method with a variety of values on 1000 glass type production data objects obtained from the central UCI Machine Learning Repository dataset. The conclusion of this research is the results of the value of K = 3 to K = 9, the best performance values obtained at K = 3, where the level of accuracy reaches 64%, 63% precision, 71% recall, and F-Measure of 67%.

Related SDGs Targets

Full Author List

Azis, Huzain; Purnawansyah; Fattah, Farniwati; Putri, Inggrianti Pratiwi, Universitas Muslim Indonesia, Urip Sumoharjo km.5, Makassar 90231, Indonesia

Abstract

Secara global, penyebab kematian tertinggi setiap tahunnya adalah penyakit kardiovaskular, yaitu penyakit yang disebabkan oleh gangguan fungsi jantung dan pembuluh darah seperti penyakit jantung koroner, gagal jantung, hipertensi, dan stroke. Penelitian ini bertujuan untuk mengukur performa metode K-Nearest Neighbor (K-NN) dan cross validation berdasarkan metrik akurasi, presisi, recall, dan f-measure pada dataset pasien kardiovaskular. Dataset yang digunakan berjumlah 1000 record dengan 11 atribut (seperti age, gender, height, dan lainnya) yang mencakup data pasien kardiovaskular dan non-kardiovaskular, serta diperoleh dari UCI Machine Learning Repository yang dikelola oleh Hungarian Institute of Cardiology Budapest. Tahapan penelitian meliputi pembagian rasio dataset menjadi 20:80, 50:50, dan 80:20, penerapan cross validation (k-fold = 10), serta proses klasifikasi menggunakan metode K-NN dengan nilai K dari 2 hingga 900. Hasil penelitian menunjukkan bahwa pada rasio 50:50 diperoleh akurasi 82%, presisi 82%, recall 82%, dan f-measure 80% pada K = 13; pada rasio 20:80 diperoleh akurasi 87%, presisi 87%, recall 97%, dan f-measure 92% pada K = 3; serta pada rasio 80:20 diperoleh akurasi 91%, presisi 92%, recall 60%, dan f-measure 72% pada K = 5.

Related SDGs Targets

Full Author List

Fattah, Farniwati; Azis,Huzain, Universitas Muslim Indonesia, Jln. Urip Sumoharjo km.05, Makassar , 90245, Indonesia

Abstract

Pengabdian kepada masyarakat dengan mitra perangkat Desa Tonasa, Kecamatan Sanrobone, Kabupaten Takalar bertujuan untuk meningkatkan literasi penggunaan dan pemanfaatan Teknologi Informasi dan Komunikasi (TIK) melalui pembuatan website desa binaan. Website (desabinaan.umi.ac.id/tonasa) dikembangkan sebagai media yang menampilkan profil Desa Tonasa serta menyajikan berbagai informasi seperti visi dan misi, struktur organisasi, jumlah penduduk, dan pengumuman lainnya. Kegiatan pengabdian ini meliputi tahapan pengumpulan data, pengambilan gambar lingkungan desa, pembuatan dan pengisian konten website, pelatihan admin, hingga pengelolaan website. Namun, dalam pelaksanaannya terdapat beberapa kendala, antara lain belum adanya staf khusus yang ditunjuk sebagai admin sehingga pengelolaan dan pembaruan konten menjadi terhambat, keterbatasan infrastruktur jaringan internet, serta kurangnya kecakapan dalam penulisan konten website. Oleh karena itu, diharapkan pihak pemangku kebijakan desa dapat menyediakan anggaran yang memadai untuk pengembangan media informasi berbasis website, sehingga ke depannya seluruh staf desa dapat berkontribusi dalam mengunggah dan memperbarui informasi pada website desa.

Related SDGs Targets

Full Author List

A'ayunnisa, Nurul; Salim, Yulita; Azis, Huzain; Universitas Muslim Indonesia, Indonesia

Abstract

Berdasarkan penelitian yang dilakukan oleh Herman dkk., peneliti mencoba mengangkat kembali metode yang diterapkan dengan menggunakan dataset yang berbeda dan dengan jumlah yang lebih banyak. Penelitian ini bertujuan untuk menghitung performa metode (akurasi, presisi, recall, dan f-measure) Gaussian Naïve Bayes. Dataset yang digunakan adalah citra tulisan tangan karakter arab. Berdasarkan hasil perhitungan performa menunjukkan tingkat akurasi tertinggi sebesar 12%, presisi 10%, recall 12%, dan f-measure 8%.

Related SDGs Targets

Full Author List

Muharram, Faturungi; Azis, Huzain; Manga, Abdul Rachman, Universitas Muslim Indonesia Fakultas Ilmu Komputer Makassar, Indonesia

Abstract

Meninjau dalam penggunaan teknologi, manusia tak pernah lepas dari kebutuhan akan sebuah informasi. Beberapa informasi dapat berupa file gambar, dokumen, dan video. Salah satu dari informasi tersebut banyak mengandung informasi penting yaitu informasi dalam bentuk file dokumen. Beberapa informasi memiliki privasi yang tidak boleh tersebar oleh public, oleh karena itu diperlukan cara dalam mengamankan informasi agar informasi tidak tersebar luas kepada pihak yang tak berwenang, dalam hal ini keamanan adalah salah satu hal yang penting. Salah satu cara yang diperlukan adalah menggunakan metode kriptografi. Dalam proses kriptografi terdapat konsep dasar yaitu enkripsi dan dekripsi. Enkripsi adalah proses dimana informasi atau data yang hendak dikirim diubah menjadi bentuk yang hampir tidak dikenali sebagai informasi awalnya dengan menggunakan algoritma tertentu. Dekripsi adalah kebalikan dari enkripsi yaitu mengubah kembali bentuk tersamar tersebut menjadi informasi awal [1]. Pada proses enkripsi terdapat beberapa cara yang dapat digunakan dan memiliki tingkat kekuatan serta kecepatan dan kelemahan dalam proses enkripsi tersendiri. Terdapat jenis model kriptografi, namun pada penelitian ini akan menyajikan analisis terhadap model kriptografi Advanced Encryption Standard (AES).

Related SDGs Targets

Full Author List

Azis, Huzain; Nirmala; Syafie, Lukman; Herman; Fattah Farniwati; Hasanuddin,Tasrif, Faculty Of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

This investigation constitutes a noteworthy progression in the advancement of more sophisticated and precise botanical image analysis. The primary objective of this inquiry is to confront the difficulties associated with the categorization of Calendula and Coreopsis flowers through the application of diverse segmentation techniques and classification algorithms. In this experiment, we employed the Canny edge detection, thresholding, mean shift, and Otsu methods to process flower images before applying Naïve Bayes, K-Nearest Neighbors, Support Vector Machine, and Decision Tree algorithms for classification. Enhanced comprehension of the integration of distinct segmentation techniques with varied classification algorithms is attained. We scrutinized accuracy, precision, recall, and F1 measure across diverse segmentation scenarios to assess the efficacy of these algorithms. Our principal discoveries consistently affirm that the Decision Tree algorithm attains the utmost accuracy levels in flower classification when coupled with mean shift segmentation, underscoring its noteworthy proficiency in this endeavor. The pivotal role of an optimal amalgamation of segmentation techniques and classification algorithms in augmenting flower recognition is underscored, thereby charting the course for subsequent investigations into the integration of diverse segmentation methods with advanced classification algorithms. This study's outcomes wield a favorable influence on the domain of botany and image analysis at large, offering support to researchers and scientists in achieving a more precise understanding and classification of plant species.

Related SDGs Targets

Full Author List

Syafie, Lukman; Satra, Ramdan; Azis, Huzain, Universitas Muslim Indonesia

Abstract

Hybrid learning model is the combination of onsite learning and online learning model. Hybrid learning model is an interesting issue in order to balance the drastically change of learning model to the learning model of ICT-based. This study was analyze the effectiveness of hybrid learning model on the student. The test were carried out before and after the learning model applied. The data was picked up from the questionnaire. The sample consisted of 40 students. By using statistical analysis, we obtain that the average difference between the pre-test and post-test scores was -29.03175. In the t-test, Ho: pre-test=post-test gives a T value=-33,890 with 39 degrees of freedom. The p-value for the two-tailed test is 0.000 less than=0.05. It was proved that the average pre-test and post-test scores are significantly different. It means the application of the hybrid model learning is effective. The implication of research is to encourage the use of e-learning technology to improve academic learning outcomes.

Related SDGs Targets

Full Author List

Fadhillah, Nurul; Azis, Huzain; Lantara, Dirgahayu, Univeritas Muslim Indonesia Fakultas Imu Komputer Makassar, Indonesia

Abstract

Untuk mengatasi kesalahan dalam pencarian kata kunci perlu dilakukan optimasi proses pencarian pada aplikasi Kamus Besar Bahasa Indonesia (KBBI) digital. Namun, tidak sedikit ditemui kesalahan dalam menuliskan kata kunci sehingga menghasilkan keluaran yang tidak sesuai dengan keinginan pengguna. Dalam hal ini diperlukan sistem yang dapat melakukan koreksi hasil pencarian kata kunci pada aplikasi KBBI digital dalam bentuk validasi hasil pencarian. Penelitian ini menggunakan metode Approximate String Matching pada algoritma Levenshtein Distance. Pada metode ini, akan diketahui jarak Levenshtein yang menjadi nilai kemiripan suatu objek bertipe string. Untuk mendapatkan nilai kemiripan dilakukan dengan menghitung jarak antar dua string dengan menghitung jumlah operasi yang terjadi seperti penambahan, penghapusan atau pengurangan karakter. Semakin rendah nilai jarak antar dua string maka semakin tinggi tingkat kemiripan kedua string tersebut dan sebaliknya. Seperti pada tingkat kemiripan antara string “varitas” dengan string “varietas” memiliki tingkat kemiripan dengan melihat Levenshtein Distance sama dengan 1 karena hanya mengalami operasi 1 kali yaitu operasi penambahan karakter dan nilai akurasi similaritas sama dengan 88 %

Related SDGs Targets

Full Author List

Manga, Abdul Rachman; Latief, Muhammad Acqmal Fadhilla; Gaffar, Andi Widya Mufila; Azis, Huzain; Satra, Ramdan; Salim,Yulita, Departement of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

Visual pattern recognition, selection of appropriate image processing techniques, and network architecture are key factors in achieving optimal model performance. This article focuses on the application of Identity Blocks in the context of image processing, especially on unbalanced datasets. Three different datasets, namely Plant Diseases, Rock & Paper Scissors, and Animal Faces, are used in this study, each with unique characteristics. Identity Block, implemented in the ResNet network architecture, helps to overcome the gradient loss problem that often occurs in deep neural networks (DNN) with deep layers. This research specifically explores Identity Block optimization using the hyperband method to improve model performance. The average performance improvement of all optimized models is 4.45% in accuracy, 5.39% in precision, 6.4% in recall, and 6.48% in F1-score. These results show that model optimization is very good at improving identity block performance using the hyperband method.

Related SDGs Targets

Full Author List

Lorem ipsum dolor sit amet consectetur adipisicing elit. Nesciunt dolorem dolores 123ad maiores doloremque ducimus distinctio ratione exercitationem dicta aperiam ipsam, officiis harum voluptate dolore pariatur?

Abstract

Lorem ipsum dolor sit amet consectetur adipisicing elit. Nesciunt dolorem dolores ad maiores doloremque ducimus distinctio ratione exercitationem dicta aperiam ipsam, officiis harum voluptate dolore pariatur?

Related SDGs Targets

Full Author List

Umar, Fitriyani; Aisyah, Universitas Muslim Indonesia

Abstract

Setiap tahunnya, di Indonesia jumlah penderita gangguan mata terus mengalami peningkatan. Beberapa jenis penyakit ini di antaranya katarak, kebutaan, kelainan refraksi dan kornea. Penyakit mata perlu mendapatkan pemeriksaan dan pengobatan dari dokter spesialis mata untuk menghindari kerusakan yang lebih parah. Seiring perkembangan teknologi, Pekerjaan yang sangat sibuk dari seorang dokter mengakibatkan bidang sistem pakar mulai dimanfaatkan untuk membantu seorang pakar/ahli dalam mendiagnosa berbagai macam penyakit. Penelitian ini membuat sistem pakar berbasis web untuk diagnosa awal penyakit mata. Sistem dibangun menggunakan bahasa PHP dengan model Unified Modelling Langguage (UML). Metode forward chaining diterapkan pada inferensi untuk merunutkan gejala-gejala yang menghasilkan kesimpulan. Adapun certainty factor untuk mengetahui berapa persen penyakit yang di derita penderita melalui hasil hitungan dalam metode tersebut. Hasil pencocokan diagnosa menunjukkan bahwa 100% sesuai antara hasil sistem dan diagnosa pakar. Sistem ini dapat menjadi sistem diagnosa awal untuk mengetahui jenis penyakit mata yang dialami tanpa konsultasi kepada spesialis mata terlebih dahulu

Related SDGs Targets

Full Author List

Jabir, Sitti Rahmah; Azis, Huzain; Widyawati, Dewi; Tenripada Andi Ulfa, Universitas Muslim Indonesia

Abstract

GRDS menghadapi kelangkaan dana, ketika diperlukan untuk merawat para korban Gaja. Gaja adalah topan bernama kelima dari musim siklon Samudra Hindia Utara 2018 yang mempengaruhi sebagian besar tempat di Tamil Nadu, India selama bulan November 2018. Tujuan dari penelitian ini adalah untuk menggunakan riwayat donasi untuk menganalisis apakah donator akan menyumbang atau tidak menggunakan regresi logistik. Data Tamil Nadu diberikan untuk menerapkan model yang dibangun untuk memprediksi donator yang paling mungkin menjadi korban topan Gaja. Pada tahap pengumpulkan data seringkali terjadi hambatan, salah satu hambatannya yaitu fenomena missing data atau data hilang. Akibat dari adanya missing data adalah pendugaan parameter menjadi tidak efisien. Ukuran data yang berkurang dapat mengakibatkan kesulitan dalam menganalisis, sehingga hasil yang didapatkan menjadi tidak valid dan tujuan dari penelitian tidak tercapai. Data yang hilang akan diisi menggunakan metode single imputation. Data yang telah diimputasi menggunakan beberapa metode akan membantu dalam melakukan prediksi. Dimana algoritma yang digunakan untuk melakukan prediksi ialah logistic regression. Beberapa data dihilangkan setelah melihat multikolinearitas. Dalam tahap pemodelan, data dibagi menjadi 2 yaitu 70% untuk data pelatihan dan 30% untuk data tes. Dimana hasil perhitungan akurasi dari model ialah 0,6129 yang menunjukkan bahwa model tidak melakukan prediksi dengan baik menggunakan metode tersebut

Related SDGs Targets

Full Author List

Azis, Huzain; Jabir, Sitti Rahmah, Universitas Muslim Indonesia

Abstract

This study focuses on the analysis of the aroma quality of tofu preserved with formalin, with the goal of developing a predictive model based on its chemical composition. Utilizing a dataset that includes various chemical components such as Hydrogen, LPG, CO, Alcohol, Propane, Methane, Smoke, and temperature, this research applies a Decision Tree model. The model is validated using 5-fold cross-validation, resulting in an accuracy of 36.79%, precision of 50.82%, recall of 36.79%, and an F1-Score of 27.58%. These results indicate the model's limitations in consistent prediction, suggesting potential improvements through other methods or the addition of variables. This study provides new insights into the relationship between chemical composition and aroma quality of formalin tofu, and opens new avenues for further research in this field.

Related SDGs Targets

Full Author List

Salim, Yulita; Puspitasari, Yustisia; Azis, Huzain; Anas, Risnayanti; Universitas Muslim Indonesia, Indonesia

Abstract

According to the World Health Organization (WHO), malocclusion is a deviation in dentofacial growth or an abnormal relationship between the teeth of both arches, which results in impaired physical function for sufferers. Causes of malocclusion include genetic factors, inappropriate growth and development processes, bad habits of children, and malnutrition. Also, malocclusion can be caused by a lack of knowledge of children, parents, and guardians of students in the school environment in maintaining oral health. Nurul Falah Kindergarten, located in Mamajang District in the middle of Makassar City. However, students in kindergarten are from the middle to lower economies with a lack of dental and oral health awareness. According to the principal, some students come with the condition of not brushing teeth and with cavities. This service activity aims to help solve the problems faced by teachers in pre-school age students by providing dental education based on Augmented Reality and Topical Application Fluor (TAF) as an effort to prevent malocclusion. It is hoped that through this activity malocclusion prevention can be done through promotive efforts on dental health. This dental extension will be complemented by the utilization of information technology advances in the form of android-based Augmented Reality (AR) technology that is able to visualize an object in 3 dimensions so that the counseling process becomes more interactive and real.

Related SDGs Targets

Full Author List

Azis, Huzain; Purnawansyah, Universitas Muslim Indonesia; Dwiyanto, Felix Andika, AGH University Of Science and Technology

Abstract

Support Vector Regression (SVR) is a supervised learning algorithm to predict continuous variable values. The basic goal of the SVR algorithm is to find the most suitable decision line. SVR has been successfully applied to several issues in time series prediction. In this research, SVR is used to predict the price of staple commodity, which are constantly changing in price at any time due to several factors making it difficult for the public to get groceries that are easy to reach. National staple commodity data consisting of 17 commodities, including shallots, honan garlic, kating garlic, medium rice, premium rice, red cayenne peppers, curly red chilies, red chili peppers, meat of broiler chicken, beef hamstrings, granulated sugar, imported soybeans, bulk cooking oil, premium packaged cooking oil, simple packaged cooking oil, broiler chicken eggs, and wheat flour. With a data set for the last 3 years, including from January 1, 2020, to December 31, 2022. There are 3 variables in the data set, namely commodity, date, and price. This research divides the entire dataset into 80% training and 20% testing data. The results of this research show that SVR using the RBF kernel produces good forecasting accuracy for all datasets with an average Mean Square Error (MSE) training data of 6,005 while data testing is 6,062, Mean Absolute Deviation (MAD) of training data is 6,730 while data testing is 6.6831, Mean Absolute Percentage Error (MAPE) training data is 0.0148 while data testing is 0.0147, and Root Mean Squared Error (RMSE) training data is 7.772 while data testing is 7.746

Related SDGs Targets

Full Author List

Azis, Huzain, Universitas Kuala Lumpur, JlnSultan Ismail, Bandar Wawasan, 50250 Kuala Lumpur, Wilayah Persekutuan Kuala Lumpur, Malaysia

Abstract

This study explores the effectiveness of Logistic Regression in predicting heart disease using a dataset derived from multiple international databases. Employing a 5-fold cross-validation method, the research aimed to evaluate the model's accuracy, precision, recall, and F1-score. Results indicated that Logistic Regression performs robustly, with accuracy ranging from 80% to 88.29%, and high recall rates, highlighting its potential as a valuable tool in medical diagnostics. Despite some variability in precision, which may lead to higher false positive rates, the model's high recall is crucial in clinical settings where missing a diagnosis can have dire consequences. The research confirmed the applicability of Logistic Regression to binary classification problems in healthcare, aligning with existing literature that supports its use in similar contexts. The study contributes to the field by demonstrating the model's consistency and reliability across diverse data subsets, reinforcing the potential for machine learning applications in healthcare diagnostics. Future research should focus on integrating Logistic Regression with other models to improve accuracy and testing the model on more current, varied datasets to enhance its generalizability and effectiveness in real-world settings.

Related SDGs Targets

Full Author List

Nasruddin, Nasruddin; Azis, Huzain; Lantara, Dirgahayu, Univeritas Muslim Indonesia Fakultas Imu Komputer Makassar, Indonesia

Abstract

Saat ini, terdapat beberapa Jenis laptop di pasaran. Dengan beragamnya jenis laptop dan spesifikasinya, membuat pengguna cukup kesulitan dalam memilih sesuai dengan kebutuhan yang di inginkan. Penelitian yang dilakukan menggunakan teknologi Augmented Reality dengan metode Markerless yang bertujuan untuk membantu pengguna memilih jenis laptop berdasarkan fisik dan merk dengan tampilan 3D (3 Dimensi). Metode Markerless mampu menangkap gambar dengan akurasi piksel Visual 3D fisik termasuk ketebalan, ukuran (inch), keyboard dan merek laptop (Acer, Toshiba, Lenovo, Dell, Asus dll). Dengan tambahan plugin atau library yang terdapat di android, dapat lebih memudahkan dalam pembuatan aplikasi Augmented Reality di dalam smartphone karena sudah tersedia tools yang bisa digunakan ketika ingin membuat teknologi Augmented Reality. Berdasarkan hasil percobaan menggunakan laptop merk asus, didapatkan hasil akurasi sebesar 88%. Untuk penelitian selanjutnya, dapat digunakan penggabungan metode Augmented Reality dengan Citra Digital agar ditemukan persentase yang lebih akurat.

Related SDGs Targets

Full Author List

Fattah, Farniwati; Azis, Huzain, Fakultas Ilmu Komputer Universitas Muslim Indonesia, Jl. Urip Sumoharjo KM. 05, Makassar 90231, Indonesia

Abstract

Pengabdian kepada masyarakat (PkM) dengan mitra Yayasan Sinergi Cendikia Indonesia bertujuan untuk membantu yayasan dalam mencapai tujuan organisasinya, khususnya sebagai lembaga yang bergerak di bidang sosial dan kemasyarakatan yang memiliki berbagai kegiatan keagamaan dan kemanusiaan sehingga membutuhkan media digital untuk mempromosikan serta menyebarkan informasi program. Sebagai yayasan yang baru berdiri, lembaga ini memerlukan media promosi yang mampu menjangkau masyarakat luas. Dalam rangka memenuhi salah satu kewajiban Tri Dharma Perguruan Tinggi, dosen Fakultas Ilmu Komputer (FIKOM) melaksanakan kegiatan pengabdian dengan memanfaatkan layanan video conference, salah satunya menggunakan aplikasi Zoom Meeting yang digunakan oleh yayasan untuk mendukung kegiatan seperti rapat, pelatihan, dan seminar. Meskipun sebagian anggota telah memahami penggunaan beberapa fitur saat meeting, masih terdapat fitur lain yang belum familiar sehingga diperlukan pendampingan, khususnya dalam penggunaan akun Zoom. Kegiatan pengabdian ini diharapkan dapat meningkatkan popularitas Yayasan Sinergi Cendikia Indonesia, khususnya di Sulawesi Selatan, serta membantu anggota yayasan dalam memanfaatkan teknologi digital untuk menjalankan tugas dan tanggung jawab secara lebih efektif.

Related SDGs Targets

Full Author List

Indra, Dolly; Satra, Ramdan; Azis, Huzain; Manga, Abdul Rachman; Universitas Muslim Indonesia, Indonesia

Abstract

Strawberry is one type of fruit that is favored by the people of Indonesia. The detection process to identify strawberries can be done by utilizing advances in computer technology, One of them is in the field of digital image processing. In this study, we made a strawberry ripeness detection system using the values of Red, Green and Blue as the reference values, while for identification in determining the type of classification using the K-Means algorithm that uses the Euclidean distance difference as the reference. Based on the results of testing using the K-Means algorithm on 51 strawberry images consisting of ripe, semi ripe and raw fruit yielding an accuracy rate of 82.14%, we also conducted tests other than strawberry images as many as 8 images yielded an accuracy rate of 100%.

Related SDGs Targets

Full Author List

Aziz, Husain; Universitas Muslim Indonesia, Indonesia

Abstract

Voice over Internet Protocol (VoIP) is a real-time service that enables voice conversations using IP networks, Steganography is the art and science of hiding secret messages in other messages that makes the secret messages are unknowable. VoIP allows as a medium cover carrier on steganography to provide the security of confidential messages. This study built by two-stage of steganography system using Least Significant Bits and Covert Channel that use image and VoIP communication as a medium in the hiding of messages. End of this study, the system will go through the process of performance and functionality testing. The results showed that the method of Covert Channel to the Payload field has the hiding of large capacity that are 56bit per packet, and by using two stages method of steganography can make a Steganogram extraction or analysis more difficult to do

Related SDGs Targets

Full Author List

Azis, Huzain; Nasrullah, Departement of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia; Abdullah Munaisyah; Ismail, Suriana, Malaysian Institute of Information Technology, Universiti Kuala Lumpur, Kuala Lumpur, Malaysia; Purnawansyah; Syafie, Lukman, , Departement of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

The use of Unmanned Aerial Vehicles (UAVs) in aerial imaging is expanding rapidly, particularly in traffic monitoring and intelligent transportation systems. Detecting small and occluded vehicles in aerial images poses significant challenges due to varying resolutions and obstructions like buildings or trees. This study seeks to enhance vehicle detection accuracy by improving You Only Look Once (YOLO) models, with a focus on small and occluded object detection. Utilizing the COWC-M dataset and advanced data augmentation techniques such as Mosaic Augmentation, this research evaluates multiple YOLO variants. The YOLOv8-L model achieved the highest mAP50 score of 0.9899, demonstrating superior detection accuracy for small objects. Additionally, the YOLOv10-L model outperformed others with the best mAP50–95 score of 0.8715, indicating strong results across different intersection-over-union (IoU) ranges. Compared to YOLO-RTUAV, which achieved an mAP50 of 0.9353, the newer YOLO models provide significant improvements in both precision and recall. These findings contribute to the development of highly efficient, real-time vehicle detection systems suitable for large-scale surveillance applications in complex environments.

Related SDGs Targets

Full Author List

Mustafa, Abdul Rahman; Belluano, Poetri Lestari Lokapitasari; Azis, Huzain, Universitas Muslim Indonesa, Makassar, Indonesia

Abstract

Aplikasi stock opname berbasis web service dirancang untuk mengatasi perbedaan pencatatan transaksi secara real-time di Perum BULOG Kanwil Papua, yang sebelumnya dilakukan secara manual menggunakan Microsoft Excel. Aplikasi ini menggunakan arsitektur Service-Oriented Architecture (SOA) yang memungkinkan integrasi data antara cabang dan kantor pusat secara efisien dan real-time. Data transaksi barang masuk dan keluar dicatat melalui sistem yang terhubung langsung ke web service, sehingga data dari seluruh cabang dapat diakses dan diperbarui secara otomatis. Hasil implementasi menunjukkan peningkatan akurasi pencatatan, efisiensi operasional, dan kemudahan aksesibilitas data, yang membantu mengurangi risiko kesalahan manusia dalam pencatatan. Pengujian sistem dilakukan melalui black box testing untuk memastikan fungsionalitas sistem berjalan sesuai kebutuhan.

Related SDGs Targets

Full Author List

Sainlia, Ahmad Fauzan; Belluano, Poetri Lestari Lokapitasari; Azis, Huzain, Program Studi Teknik Informatika, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

Dalam hal pengaturan jadwal dakwah, metode PIECES digunakan untuk mengatur sinkronisasi jadwal, durasi dakwah, memberikan notifikasi ketika mubaligh sudah memasuki waktu dakwah dan mengatur jadwal dakwah para mubaligh dengan baik. Analisa kinerja dalam PIECES yaitu menyediakan sistem yang dapat mengatur jadwal mubaligh, dalam analisa informasi yaitu memberikan informasi ke mubaligh ketika mendekati waktu dakwahnya, dalam analisa ekonomi memberikan informasi mengenai transparansi honorarium pada mubaligh, dalam analisa efisiensi memberikan durasi waktu bagi para mubaligh dan dalam analisa pelayanan memberikan notifikasi ketika mubaligh sudah memasuki waktu dakwah. Aplikasi Penjadwalan Dakwah Mubaligh menggunakan Metode PIECES dirancang untuk dapat membantu petugas masjid dalam menyusun jadwal para mubaligh serta memberikan informasi terkait durasi dakwah bagi tiap mubaligh sehingga para mubaligh tidak lupa dengan jadwal dakwah mereka serta dapat mengatur tranparansi pendapatan tiap mubaligh. Sistem aplikasi berbasis mobile dapat diakses melalui perangkat android, dimana sistem dilengkapi notifikasi aktif yang dapat diakses setiap saat oleh para pengguna sesuai kebutuhan, sehingga mubaliqh dapat mengatur durasi dakwah dengan jelas, dan tidak ada lagi jadwal yang berbentura antar mubaligh satu dengan lainnya.

Related SDGs Targets

Full Author List

H, Sarah Fila Nurul Fitri; Fattah, Farniwati; Azis, Huzain, Universitas Muslim Indonesia

Abstract

Machine learning is an approach in computer science where systems or models can learn from data and experience to improve performance or perform specific tasks. There are several popular machine learning algorithms, such as naïve bayes, decision tree, K-NN, and SVM. This study aims to compare the performance of accuracy, precision, recall, and F-1 score in sentiment analysis of body shaming topics on Social Media X (formerly known as Twitter) by applying decision tree, K-NN, and SVM methods and identifying the most effective algorithm in classifying the data. Based on the classification performance testing results, it can be concluded that the classification method using the trigram feature model provides the best performance compared to other methods. The trigram model is able to achieve high recall, particularly in recognizing positive classes, without significantly compromising accuracy

Related SDGs Targets

Full Author List

Fattah, Farniwati; Maharani, Aulia; Azis, Huzain, Fakultas Ilmu Komputer, Universitas Muslim Indonesia

Abstract

Penggunaan kartu magnetic stripepada wahana permainan rentan terhadap akses yang tidak sah, seperti skimming, yang dapat merugikan pengelola dan penyedia wahana. Penetration testingmerupakan metode yang dapat digunakan untuk mengidentifikasi dan eksploitasi kerentanan. Pada pengujian penetration testingterdapat tujuh fase yang digunakan yaitu pre-engagement, information gathering, threat modeling, vulnerability analysis, exploitation, post exploitation, danreporting. Dalam Penelitian sebelumnya menunjukkan bahwa komunikasi antara magnetic stripe readerdan komputer utama dilakukan melalui koneksi kabel, yang menghasilkan layanan confidentiality danavailability. Namun, pada penelitian ini, pengimplementasian penetration testing menggunakan koneksi nirkabel menghasilkan temuan bahwa layanan keamanan yang tersedia adalah availability. Oleh karena itu, dapat disimpulkan bahwa komunikasi baik melalui koneksi kabel maupun nirkabel tidak terdapatlayanan keamanan integrity. Rekomendasi bagi penyedia layananuntuk meningkatkan kemanan kartu di lokasi tersebut dengan menerapkan enkripsi data.

Related SDGs Targets

Full Author List

Manga, Abdul Rachman; Nirmala; Azis, Huzain; Fattah, Farniwati; Salim, Yulita; Darwis, Herdianti, Departement of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

Flower image classification poses a challenge in digital image processing, requiring effective methods for feature extraction and classification. The aim of this research is to improve the accuracy of flower image classification by employing ResNet-50 for feature extraction, with and without segmentation, and evaluating the effectiveness of various classification algorithms. The dataset consists of images of Calendula and Coreopsis flowers, totaling 2,025 samples. Four segmentation techniques-Canny, Thresholding, Otsu, and Mean Shift-along with a non-segmentation approach are applied. The features extracted using ResNet-50 are classified with Naive Bayes, Support Vector Machine (SVM), Decision Tree, and K-Neighbors Classifier (KNN). Performance evaluation is conducted using accuracy, precision, recall, and F1-score with 5-fold cross-validation. The results show that SVM delivers the best performance in most scenarios. The highest accuracy in segmentation scenarios was achieved with the Mean Shift technique at 0.91, while the non-segmentation approach yielded the highest accuracy of 0.95. The non-segmentation approach proves more effective, indicating segmentation is not always required for high classification accuracy with ResNet-50. This study shows that ResNet-50, especially without segmentation, can significantly improve flower image classification compared to traditional methods, opening opportunities for more efficient systems.

Related SDGs Targets

Full Author List

Harlinda; Rendi, Ahmad; Azis, Huzain; Indra, Dolly; Hayati, Lilis Nur; Kurniati, Nia, Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

The Cia-cia script, one of Indonesia's threatened cultural heritages, was chosen as the object of study due to the lack of research and documentation on the script. This study aims to create an automated system for classifying Cia-cia letters by utilizing the MobileNetV2 architecture and Convolutional Neural Networks (CNN). By applying deep learning techniques, the developed system achieves high accuracy, reaching 98.35% after 100 epochs. The research involved collecting 1823 handwriting samples covering 23 Ciacia alphabets, processed through a series of augmentation and normalization techniques to improve model performance. The results show that artificial intelligence-based technology is effective in documenting and preserving traditional scripts, while providing a foundation for the development of educational applications that can reintroduce Cia-cia alphabets to the younger generation. This research contributes to cultural preservation by integrating modern technology to ensure the continued use of the Cia-cia script in the digital era.

Related SDGs Targets

Full Author List

Manga, Abdul Rachman; Utami, Aulia Putri; Azis, Huzain; Salim, Yulita; Faradibah, Aulia, Department of Computer Engineering, Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

The surge in machine learning (ML) and artificial intelligence has revolutionized medical diagnosis, utilizing data from chest ct-scans, COVID-19, lung cancer, brain tumor, and alzheimer parkinson diseases. However, the intricate nature of medical data necessitates robust classification models. This study compares support vector machine (SVM), naïve Bayes, k-nearest neighbors (K-NN), artificial neural networks (ANN), and stochastic gradient descent on multi-class medical datasets, employing data collection, Canny image segmentation, hu moment feature extraction, and oversampling/under-sampling for data balancing. Classification algorithms are assessed via 5-fold cross-validation for accuracy, precision, recall, and F-measure. Results indicate variable model performance depending on datasets and sampling strategies. SVM, K-NN, ANN, and SGD demonstrate superior performance on specific datasets, achieving accuracies between 0.49 to 0.57. Conversely, naïve Bayes exhibits limitations, achieving precision levels of 0.46 to 0.47 on certain datasets. The efficacy of oversampling and under-sampling techniques in improving classification accuracy varies inconsistently. These findings aid medical practitioners and researchers in selecting suitable models for diagnostic applications.

Related SDGs Targets

Full Author List

Syafie, Lukman; Azis, Huzain; Admojo, Fadhilah Tangguh, Universiti Kuala Lumpur, 50250 Kuala Lumpur, Malaysia

Abstract

Introduction: The preservation of Javanese script as part of Indonesia’s cultural heritage is increasingly urgent in the digital era, especially due to declining literacy among younger generations. This study aims to explore the effectiveness of YOLOv8, an advanced object detection algorithm, for recognizing handwritten Javanese numerals to support efforts in cultural digitization and education. Methods: A dataset of 2,790 handwritten Javanese numerals (0–9) was collected from 93 respondents. Each numeral was manually annotated using bounding boxes via the MakeSense.ai platform. The YOLOv8 model was trained using 80% of the data and validated on the remaining 20%. Training was performed in the PyTorch framework with data augmentation techniques to increase robustness. Model performance was evaluated using precision, recall, F1-score, and mean Average Precision (mAP), along with visualization through confidence curves and confusion matrices. Results: The model achieved a high validation precision of 88.3%, recall of 89.1%, and mAP of 0.88 at IoU 0.90. F1-score peaked at a confidence threshold of 0.89, while certain numerals like 'six' and 'nine' achieved near-perfect detection. Visualizations confirmed the model’s ability to accurately classify and localize characters in both training and unseen data. Minor misclassifications occurred between visually similar numerals. Conclusions: YOLOv8 demonstrates high effectiveness in recognizing handwritten Javanese numerals and holds significant potential for digital heritage preservation. Future work should focus on expanding the dataset, improving generalization under varied conditions, and integrating this model into educational tools and augmented reality applications for interactive learning.

Related SDGs Targets

Full Author List

Jabir, Sitti Rahmah; Azis, Huzain; Mansyur, St. Hajrah, Universitas Muslim Indonesia

Abstract

One of the key indicators of the growth of a college is the number of students that are enrolled in the institution on an annual basis. Student enrollment is a crucial element in the growth of a college, particularly in the case of private institutions. When examining students' aspirations for higher education, several studies use data mining techniques to forecast the interests of students who will pursue college. Researchers adopt various ways to extract valuable information from data. Prior research has shown that the decision tree technique outperforms alternative methods. The random forest, in addition to the decision tree, is often used for predicting data mining tasks. Given the above background information, the author will conduct a study titled "Comparative Analysis of Decision Tree and Random Forest Algorithms in Predicting College Interests." According to the study findings, the decision tree outperforms the random forest in terms of outcomes. The accuracy of the decision tree model is 0.81, whereas the accuracy of the Random Forest model is 0.74. All in all, the Decision Tree approach will be used as the ultimate outcome for the implementation of Business Analytics.

Related SDGs Targets

Full Author List

Salim, Yulita; Manga', Abdul Rachman; Azis, Huzain; Syafie, Lukman, Universitas Muslim Indonesia

Abstract

Universitas Muslim Indonesia (UMI) has developed several applications for managing the campus's digital information and management systems, both internally and externally. However, several applications were previously created in the development of information system applications at UMI. However, these applications were not well-suited for long-term use due to their complexity and lack of integration. Therefore, UMI aims to create a fully integrated and well-managed campus information system by implementing the concept of microservices. The microservices approach involves dividing large applications into smaller interconnected components. This approach facilitates the management of application systems and enables better integration. Moreover, the microservices approach simplifies system maintenance for application developers, as each application is separated into smaller components

Related SDGs Targets

Full Author List

Belluano, Poetri Lestari Lokapitasari; Huzain Azis, Universitas Muslim Indonesia

Abstract

Arsitektur microservice memecah sistem yang kompleks dan besar menjadi serangkaian layanan kecil dan mandiri. Salah satu pola arsitektur yang umum digunakan adalah pendekatan event driven, yang memungkinkan komunikasi berbasis event antar layanan. Namun, pendekatan ini juga membawa risiko kehilangan data, yang dapat diatasi dengan pola orkestrasi menggunakan Apache Kafka sebagai message broker. Kafka menyediakan platform yang cocok untuk komunikasi event driven dengan kemampuannya dalam menyimpan, menerima, dan mengirim pesan secara asinkron. Penelitian ini bertujuan membangun aplikasi e-ticketing berbasis web dengan menggunakan arsitektur microservice dan Kafka. Hasilnya adalah sebuah aplikasi e-ticketing yang menggunakan Kafka untuk komunikasi antar layanan, dengan implementasi lima topik untuk proses transaksi antar gate-ticketing-service dan gate-acl-service secara asinkron menggunakan Kafka sebagai media pengiriman event.

Related SDGs Targets

Full Author List

Herdianti; Manga, Abdul Rachman; Azis, Huzain, Universitas Muslim Indonesia

Abstract

Sistem informasi berbasis web telah menjadi salah satu trend pemanfaatan teknologi yang saat ini ditemui di dunia pariwisata dan ekonomi kreatif sebagai salah satu alternatif penyampaian informasi (pengetahuan dan berita) berbasis Internet yang dikeluarkan oleh dinas pariwisata atau pengelola tempat wisata dengan tujuan untuk kemudahan dalam pendistribusian informasi pada publik, baik masyarakat sekitar, wisatawan domestik, maupun wisatawan mancanegara. Sehubungan dengan hal tersebut, Fakultas Ilmu komputer sebagai lembaga yang berperan aktif dalam pengembangan dan penerapan teknologi informasi di Kawasan Indonesia Timur dengan visi fakultas yaitu “Smart Village” yang bersinergi dengan visi Universitas Muslim Indonesia terkait “Halal Issues” melakukan peningkatan kualitas dan pengembangan desa di Lembang Marinding Kecamatan Mengkendek Tana Toraja khusus di bidang pariwisata halal dan ekonomi kreatif melalui sebuah kegiatan pengabdian kepada masyarakat (PkM) yang dilaksanakan dalam bentuk pengembangan website bagi pengelola Gunung Kandora Kabupaten Tana Toraja demi peningkatan pendapatan ekonomi daerah dan secara khusus untuk kesejahteraan masyarakat sekitar

Related SDGs Targets

Full Author List

Azis, Huzain; Jabir, Sitti Rahmah, Universitas Muslim Indonesia

Abstract

Salah satu penerapan bidang teknologi yang berkembang ialah multimedia. Dalam multimedia terdapat Animasi 3 dimensi (3D), yang mana merupakan proses pembuatan pergerakan gambar dalam lingkaran 3 dimensi. Pada animasi 3D, sebuah perangkat lunak menciptakan real virtual dalam 3 dimensi dan perubahan (Gerakan) dihitung 3 aksis (x, y dan z). berdasarkan hasil tersebut, didapatkan sebuah objek yang dapat terlihat dari sisi tampak muka, samping, belakang, atas, dan bawah. Dimana orang dapat menjelajahi objek tersebut dari sudut pandang manapun. Tim Program Kemitraan Program Kemitraan Masyarakat (PKM) Pemula UMI memalukan riset pada desa Marinding, yang ditemukan belum terdapat penerapan aset 3D pada rumah Tongkonan yang ada di Desa Marinding). Dimana budaya pada desa tersebut yang masih sangat kental yang menjadikan sebagai objek budaya yang patut untuk dilestarikan. Aset 3D yang dapat diterapkan pada rumah Tongkonan nantinya dapat membantu masyarakat khususnya desa Marinding dalam memperlihatkan objek rumah Tongkonan mereka. Berdasarkan permasalahan yang telah dipaparkan sebelumnya, PKM Pemula UMI ingin memberikan solusi pelestarian budaya dengan berpatisipasi dalam melestarikan kebudayaan Tana Toraja dalam hal implementasi 3D aset pada salah satu objek budaya yang ada disana yaitu rumah Tongkonan. Desa yang akan dipilih untuk dilakukan pengabdian yaitu pada Desa Marinding. Implementasi Aset 3D yang nantinya akan dibuat diharapkan dapat membatu masyarakat dalam melestarikan budaya menerapkan aset 3D pada rumah Tongkonan.Hasil yang diperoleh dari kegiatan pengabdian yaitu mitra kantor desa Tamangapa dapat memanfaatkan aset 3D dalam pelestarian budaya, mitra Kantor Desa Tamangapa dapat membantu wisatawan dalam mengenalkan objek budaya mereka khusunya rumah Tongkonan, wisatawan dapat dengan mudah mempelajari terkait rumah tongkonan yang ada di Kab. Tana Toraja dan perangkat desa dapat memperlihatkan dengan mudah objek wisata yang dimiki dalam bentuk 3D. Sebagai penunjang dalam kegiatan tersebut, staf yang mengikuti kegiatan pelatihan diberikan modul implementasi aet 3D rumah Tongkonan pada Desa Marinding. Luaran akhir berupa publikasi di media masssa (inipasti.com) dan artikel yang akan di publikasipadan pada jurnal ILKOMAS (Ilmu Komputer Untuk Masyarakat (ILKOMAS)

Related SDGs Targets

Full Author List

Suzanti, Ika Oktavia; Kamil, Fajrul Ihsan; Rochman, Eka Mala Sari; Rachman, Fika Hastarita; Solihin, Firdaus, Informatics Engineering Department, University of Trunojoyo Madura, Bangkalan, Indonesia; Huzain Azis, Malaysian Institute of Information Technology (MIIT), Universiti Kuala Lumpur, Kuala Lumpur, Malaysia; Suni, Alfa Faridh, Computer Science Department, Newcastle University, United Kingdom

Abstract

Hidden paradise is a term that aptly describes the island of Madura, which offers diverse tourism potential. Through the Google Maps application, tourists can access sentiment-based information about various attractions in Madura, serving both as a reference before visiting and as evaluation material for the local government. The Multinomial Naïve Bayes method is used for text classification due to its simplicity and effectiveness in handling text mining tasks. The sentiment classification is divided into three categories: positive, negative, and mixed. Initial analysis revealed an imbalance in sentiment data, with most reviews being positive. To address this, sampling techniques—both oversampling and undersampling—were applied to achieve a more balanced data distribution. Additionally, the Adaptive Boosting ensemble method was used to enhance the accuracy of the Multinomial Naïve Bayes model. The dataset was split into training and testing sets using ratios of 60:40, 70:30, and 80:20 to evaluate the model’s stability and reliability. The results showed that the highest F1-score, 84.1%, was achieved using the Multinomial Naïve Bayes method with Adaptive Boosting, which outperformed the model without boosting, which had an accuracy of 76%.

Related SDGs Targets

Full Author List

As'ad, Ihwana, Program StudiSistem Informasi, Universitas Muslim Indonesia, Makassar, Indonesia; Salim, Yulita; Azis, Huzain, Program StudiTeknik Informatika, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

Anak-anak dan remaja sering menggunakan media sosial dan bermain game secara berlebihan, yang dapat berdampak negatif terhadap kesehatan dan prestasi akademik. Kurangnya kesadaran akan keamanan digital dan etika penggunaan teknologi menjadikan mereka rentan terhadap ancaman siber. Kegiatan pengabdian ini bertujuan untuk meningkatkan literasi digital siswa Sekolah Kebangsaan Syeikh Mohd Idris Al-Marbawi di Malaysia melalui workshop interaktif. Metode pelaksanaan meliputi analisis kebutuhan, penyusunan dan pelaksanaan materi pelatihan, kepada 20 siswa kelas 5. Hasil kegiatan menunjukkan peningkatan pemahaman siswa terhadap penggunaan gadget yang bijak dan etika digital. Kegiatan ini juga mendorong pengembangan kemampuan interpersonal dan kesadaran terhadap keamanan siber.

Related SDGs Targets

Full Author List

Nanda, Sultan Aziz Syaifullah, rogram Studi Teknik Informatika,Universitas Muslim Indonesia, Jalan Urip Sumoharjo, Makassar 90231, Indonesia; Azis, Huzain, Universiti Kuala Lumpur, 50250 Kuala Lumpur, Malaysia

Abstract

This research aims to design and develop a power control and monitoring device based on NodeMCU and the PZEM-004T module as the central controller and monitoring unit, enabling users to remotely control and monitor household lamp power consumption. The method used in this study is an experimental approach to evaluate the performance of the developed lamp power control and monitoring device. The NodeMCU ESP8266 microcontroller is employed to execute program instructions and enable remote control, while the PZEM-004T module functions as the power consumption meter. Through serial communication, it provides real-time data on power (Watt), energy consumption (kWh), and cost (IDR) to a smartphone via the Thinger.io web platform. Users can manage the ON-OFF status of home lighting and monitor power usage in real-time. The testing results indicate that the developed prototype successfully supports real-time control and monitoring with an average power measurement error of only 0.10%.

Related SDGs Targets

Full Author List

Latip, Abdul, Institute of Advanced Informatics and Computing, Tasikmalaya 46115, Indonesia; Azis, Huzain, Malaysian Institute of Information Technology, Universiti Kuala Lumpur, Malaysia; Himawan, Hidayatulloh, Department of Informatics, UPN Veteran Yogyakarta, Indonesia, Faculty of Information and Communication Technology, Universiti Teknikal Malaysia, Malaysia; Kurnia, Dian Ade, Department of Informatics Management, STMIK IKMI Cirebon, Indonesia, Faculty of Information and Communication Technology, Universiti Teknikal Malaysia, Malaysia

Abstract

Cirrhosis is a chronic liver disease with significant health implications, responsible for 56,585 deaths annually, and ranking as the 9th leading cause of mortality worldwide. Early detection is crucial for effective treatment and better patient outcomes, as cirrhosis can progress to irreversible damage if not addressed in its initial stages. This research focuses on developing an advanced, integrated method for detecting cirrhosis by employing a combination of Synthetic Minority Over-sampling Technique (SMOTE) and machine learning models, specifically Gradient Boosting and XGBoost. The use of SMOTE is critical in this study as it addresses class imbalance in the dataset, which is a common challenge in medical diagnosis problems, especially when dealing with rare or minority conditions like cirrhosis. Class imbalance can lead to biased models that perform poorly on the minority class, which, in this case, could mean missing crucial cirrhosis diagnoses. SMOTE oversamples the minority class to ensure a more balanced dataset, which improves the model's ability to detect cirrhosis accurately. The research further includes a performance comparison between two powerful machine learning algorithms: Gradient Boosting and XGBoost. Gradient Boosting is known for its ability to optimize the model by focusing on misclassified instances in a sequential manner, while XGBoost, an advanced version of Gradient Boosting, is renowned for its speed and efficiency due to parallel processing and advanced regularization techniques.

Related SDGs Targets

Full Author List

Saly, Intan Novita; Purnawansyah, Universitas Muslim Indonesia, Sulawesi Selatan, Makassar, 90231, Indonesia; Azis, Huzain, Universitas Muslim Indonesia, Sulawesi Selatan, Makassar, 90231, Indonesia, Universiti Kuala Lumpur, Kuala Lumpur, Malaysia, 50250, Malaysia

Abstract

This study aims to evaluate the performance of various classification methods in recognizing handwritten Arabic numerals, particularly the K-Nearest Neighbors (KNN), Gaussian Naive Bayes (GNB), and NU Support Vector Classifier (NU SVC) algorithms. In this study, a dataset of handwritten Arabic numerals consisting of 9,350 samples with 10 different classes was used. The research process involved data collection, data labeling, dividing the dataset into training and testing data, implementing classification algorithms, and performance testing using cross-validation methods. The results showed that NU SVC had more stable performance with accuracy close to KNN, while GNB showed the lowest performance. The conclusion of this study emphasizes that the selection of algorithms and parameter optimization is crucial to improve the accuracy and efficiency of handwriting recognition systems. Support Vector Machine (SVM) based algorithms proved to be superior in handling complex classification tasks compared to GNB. This study provides significant contributions to the field of handwriting recognition, particularly in the context of Arabic numeral handwriting, and can serve as a reference for developers of optical character recognition (OCR) systems in the future. Future research is recommended to increase the variety of datasets and further explore parameter optimization and data preprocessing techniques to improve system accuracy.

Related SDGs Targets

Full Author List

Bustam, Faida Daeng; Purnawansyah, Universitas Muslim Indonesia, Sulawesi Selatan, Makassar, 90231, Indonesia; Azis, Huzain, Universitas Muslim Indonesia, Sulawesi Selatan, Makassar, 90231, Indonesia, Universiti Kuala Lumpur, Kuala Lumpur, Malaysia, 50250, Malaysia

Abstract

This research explores the performance of several classification algorithms on handwritten Lontara digits, a script traditionally used by the Bugis and Makassar communities in South Sulawesi, Indonesia. The dataset comprises 10,890-digit samples, contributed by 99 individuals, and is categorized into 10 distinct classes corresponding to the digits 0- 9. The classification methods evaluated in this study include K-Nearest Neighbors (KNN), Gaussian Naive Bayes (GNB), and Nu-Support Vector Classifier (NuSVC). Cross-validation techniques are employed to evaluate the performance of these classifiers using standard metrics such as accuracy, precision, recall, and F1 score. The findings demonstrate varying levels of performance across the algorithms. Notably, GNB achieves the highest recall, indicating its ability to correctly identify positive samples, whereas KNN and NuSVC exhibit moderate effectiveness across other performance metrics. KNN shows potential with its simple yet robust approach to classifying complex datasets, while NuSVC demonstrates a balanced performance, particularly in precision. However, all classifiers face challenges in achieving optimal accuracy, particularly due to the complexity of the handwritten Lontara digits, which exhibit unique and intricate patterns. The study concludes by suggesting that further improvements can be achieved by refining feature extraction techniques and optimizing the classifiers used. Enhancing feature extraction could provide better representations of the Lontara digits, potentially leading to improved classification accuracy. Additionally, algorithm optimization and the exploration of more advanced classification methods could further enhance the overall performance. This research provides a foundation for the development of automated recognition systems for Lontara script, contributing to its preservation and modern use.

Related SDGs Targets

Full Author List

Salim, Yulita; Azis, Huzain; Darwis, Herdianti; Purnawansyah,Departement of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia; Kurubacak, Gulsun; Anggreani, Desi;

Abstract

The scope of this book focuses on how information technology may assist in achieving goals and in providing solutions to problems such as a pandemic. Research on the Internet and on technology has been done, and the findings have applications in various sectors that rely on interdisciplinary knowledge. This book explores and describes state-of-the-art research conducted during the COVID-19 pandemic. Topics covered include the IT viewpoint and the rules governing digital transformation throughout the pandemic. The Digital Revolution sped up by a decade during COVID-19, which impacted both the user experience and that of software developers. As a component of the digital transformation process, this book explores the experiences of both the user and developer when attempting to change and adapt while utilizing an information technology program. This book includes five topics: (1) multidisciplinary artificial intelligence, (2) Smart City and Internet of Things applications, (3) game technology and multimedia applications, (4) data science and business intelligence, and (5) IT hospitality and information systems. Each topic is covered in several book chapters with some application in several countries, especially developing countries. The chapters provide insight from contributors with different perspectives and several diverse fields who present new ideas and approaches to solving problems associated with the worldwide pandemic.

Related SDGs Targets

Full Author List

Indra, Dolly; Umar, Fitriyani; Fattah, Farniwati; Azis, Huzain; Manga, Abdul Rachman, Faculty Of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

The global spread of COVID-19 had altered human behavior. One example was the shift from direct touch to less contact in interpersonal interactions. At that time, during the COVID-19 pandemic, digital technology was vital for reducing and eliminating social, physical, and psychological risk factors and managing the long-term consequences of social isolation and lockdown loneliness. Throughout the previous decade, various nations, notably developing nations, have embraced technology and adapted it to local conditions in response to the pandemic. The technologies are advantageous and might be expanded for further applications. This chapter will discuss deploying various technologies, including an automatic barrier gate, a smart stick for the blind, and automatic handwashing. These instruments utilized microcontroller technology. These tools are helpful, but they require further improvement.

Related SDGs Targets

Full Author List

Azis, Huzain; Bustam, Faida Daeng, Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

This dataset contains images of handwritten Lontara numerals ranging from 0 to 9. It comprises a total of 10890 samples, with 1089 images for each numeral class. The images were collected from various individuals to ensure diversity in handwriting styles. Key Features: Classes: 10 (Lontara numerals 0-9) Total Samples: 10890 Samples per Class: 1089 Image Format: Grayscale Data Collection and Labeling: The dataset was created by collecting handwritten numerals from participants with different handwriting styles. Each image was manually labeled to ensure accurate and consistent annotations. The data collection and labeling process was meticulously carried out by one of the authors. Usage: This dataset is suitable for training and testing machine learning models for handwritten numeral recognition. It can be used in various applications such as optical character recognition (OCR) systems, pattern recognition, and other related fields. Contributors: Author 1: Conducted the data collection and labeling process, ensuring accurate and consistent annotations for all samples. Author 2: Handled the data preprocessing, including image normalization and augmentation. Author 3: Developed the script for data collection and managed the overall project coordination. Author 4: Performed the quality check and validation of the dataset. Acknowledgments: We would like to thank all the participants who contributed their handwritten numerals for this dataset. License: CC BY NC 3.0 You are free to adapt, copy or redistribute the material, providing you attribute appropriately and do not use the material for commercial purposes.

Related SDGs Targets

Full Author List

Azis, Huzain; Saly, Intan Novita, Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

This dataset contains images of handwritten Arabic numerals ranging from 0 to 9. It comprises a total of 9350 samples, with 935 images for each numeral class. The images were collected from various individuals to ensure diversity in handwriting styles. Key Features: Classes: 10 (Arabic numerals 0-9) Total Samples: 9350 Samples per Class: 935 Image Format: Grayscale Image Size: 28x28 pixels (adjust if different) Data Collection and Labeling: The dataset was created by collecting handwritten numerals from participants with different handwriting styles. Each image was manually labeled to ensure accurate and consistent annotations. The data collection and labeling process was meticulously carried out by one of the authors. Usage: This dataset is suitable for training and testing machine learning models for handwritten digit recognition. It can be used in various applications such as optical character recognition (OCR) systems, pattern recognition, and other related fields. Contributors: Author 1: Conducted the data collection and labeling process, ensuring accurate and consistent annotations for all samples. Author 2: Handled the data labelling process. Acknowledgments: We would like to thank all the participants who contributed their handwritten numerals for this dataset. License: CC BY NC 3.0 You are free to adapt, copy or redistribute the material, providing you attribute appropriately and do not use the material for commercial purposes.

Related SDGs Targets

Full Author List

Bahri, Nur; Salim, Yulita; Azis, Huzain; Universitas Muslim Indonesia, Indonesia

Abstract

Dinas Perhubungan Kota Kendari menjadi salah satu kota yang telah menerapkan teknologi ATCS. Proses pemantau dilakukan menggunakan CCTV melalui jaringan internet yang dipantau secara real time melalui ruang kontrol Dinas Perhubungan Kota Kendari. Penerapan layanan video surveilance ATCS pada dinas perhubungan kota Kendari masih sering terjadi kendala seperti akses video surveillance yang dilakukan secara real-time mengalami buffering sehingga kualitas video yang ditampilkan tidak optimal. Permasalahan yang terjadi tersebut perlu dilakukan tindak lanjut penanganan dengan melakukan analisa layanan atau yang dikenal dengan Quality of Service. untuk menentukan apakah kualitas jaringan pada Layanan Video surveillance ATCS yang digunakan telah sesuai atau perlu dilakukan peningkatan kualitas sesuai standarisasi Tiphon dengan menggunakan metode Action Research (AR). Hasil penelitian menunjukkan hasil dari penguuran jaringan dinas Perhubungan Kota Kendari mendapatkan nilai QoS “3,55” dengan indeks “memuaskan” dan Pada Provider data (Tri) dengan nilai QoS “3,31” dengan kategori “memuaskan” yang telah di kategorikan pada standarisasi Tiphon.

Related SDGs Targets

Full Author List

Purnawansyah, Department of Electrical Engineering and Informatics, Universitas Negeri Malang, Jl. Semarang, Malang, 65145, Indonesia; Department of Computer Science, Universitas Muslim Indonesia, Jl. Urip Sumoharjo, Makassar, 90231, Indonesia; Darwis, Herdianti; Azis, Huzain, Department of Computer Science, Universitas Muslim Indonesia, Jl. Urip Sumoharjo, Makassar, 90231, Indonesia; Wibawa, Aji Prasetya; Widiyaningtyas, Triyanna, Department of Electrical Engineering and Informatics, Universitas Negeri Malang, Jl. Semarang, Malang, 65145, Indonesia; Haviluddin, Department of Computer Science, Universitas Mulawarman Jl. Kuaro Gunung Kelua, Samarinda, 75119, Indonesia

Abstract

This study aims to assess the effectiveness of various algorithms in the realms of supervised and semi-supervised learning applied to three multiclass facial image datasets: JAFFE, Georgia tech, and Yale. The datasets were partitioned into proportions of 80:20, 75:25, and 50:50 for supervised learning, while semi-supervised learning was conducted with labelled and unlabeled data ratios of 20:80, 25:75, and 50:50. The evaluated algorithms include convolutional neural networks (CNNs), decision tree, long short-term memory, K-nearest neighbors (K-NNs), multilayer perceptron, and support vector classifier (SVC), each with varying parameters. Experimental outcomes reveal that the performance of models depends on the dataset partitioning strategies and the type of algorithms used. Specifically, linear and polynomial SVC consistently yield favorable results in supervised learning, particularly demonstrating efficacy on the Georgia tech dataset. Conversely, on the JAFFE and Yale dataset, linear SVC and K-NN emerge as optimal choices. The inclusion of semi-supervised learning enhances insights, particularly evident in the Georgia tech dataset, where the combination of labeled and unlabeled data significantly improves accuracy, especially when leveraging linear SVC algorithm. Although there are some instances of sub-optimal performance in certain algorithms like CNN on specific datasets, this research provides comprehensive insights into the effectiveness of various models in contexts of limited-label learning. The implications of these findings are crucial in advancing the development of adaptive and robust facial recognition systems, especially in navigating datasets characterized by diverse variations and complexities.

Related SDGs Targets

Full Author List

Harlinda, L; Rahma, Dewi Ernita; Azis, Huzain; Irawati; Ramdaniah, Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia; Isnan, Mahmud, Computer Science Department School of Computer Science Bina Nusantara University, Jakarta, Indonesia

Abstract

The objective of this work is to develop an automatic classification system that can accurately and consistently distinguish between healthy and Curvularia-infected oil palm leaves, even in the face of inadequate data. The proposed approach uses the ResNet50 model as a feature extractor in conjunction with more conventional machine learning techniques like Random Forest and K-Nearest Neighbors. With a focus on implementation feasibility and efficiency in real-world applications, this combination has shown competitive performance when compared to the end-to-end ResNet50 approach. With these contributions, this study offers methods for diagnosing plant diseases that are intelligent, flexible, and resource-efficient, facilitating digital transformation in the agriculture industry. These findings highlight the potential of hybrid deep learning systems in providing accurate, fast, and cost-effective disease diagnosis solutions for oil palm plantations, particularly in tropical environments with limited resources.

Related SDGs Targets

Full Author List

Mubaraka, Muhammad Afdhal; Fattah, Farniwati; Azis, Huzain, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

Perkembangan teknologi Internet of Things (IoT) telah membawa dampak signifikan dalam kehidupan sehari-hari, terutama dalam pengelolaan rumah. Tugas akhir ini memiliki tujuan untuk merancang dan membangun prototipe smart home berbasis ESP32 yang dilengkapi dengan fitur sistem keamanan pintu, lampu pintar, dan AC pintar. ESP32 dipilih sebagai mikrokontroler karena kemampuannya dalam koneksi Wi-Fi dan Bluetooth, serta performanya yang baik dalam aplikasi IoT. Penelitian ini menggunakan metode perancangan prototipe yang mencakup analisis kebutuhan dan perancangan sistem. Hasil dari perancangan ini menunjukkan bahwa prototipe yang dibangun dapat berfungsi dengan baik dalam mengontrol akses pintu, pencahayaan, dan suhu ruangan secara efisien. Perancangan ini juga menunjukkan bahwa sistem dapat dioperasikan dengan tingkat respons yang cepat, yaitu kurang dari 2 detik untuk setiap perintah yang diberikan. Dengan demikian, penelitian ini tidak hanya memberikan solusi praktis untuk meningkatkan kenyamanan dan keamanan rumah, tetapi juga berkontribusi terhadap pengembangan teknologi smart home di Indonesia

Related SDGs Targets

Full Author List

Azis, Huzain; M, Muh Fatwah Fajriansyah; Darwis, Herdianti; Purnawansyah; Hasanuddin, Tasrif; Sugiarti, Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

Automatic Speech Recognition (ASR) systems perform well for high-resource languages but remain unreliable for regional dialects with limited data. The Makassar dialect, spoken by approximately nine million people in South Sulawesi, Indonesia, presents unique phonetic and grammatical challenges, including high-frequency particles such as mi, ji, and ko. This study introduces the first benchmark evaluation of state-of-the-art ASR models on Makassar speech. Four models, Whisper (tiny, base, small) and Wav2Vec2 Large XLSR Indonesian, were tested on 305 spontaneous utterances (∼10 minutes) from 10 native speakers. Results show severe performance degradation: the best model (Wav2Vec2 Indonesian) reached 87.73% of word error rate (12.27% accuracy). Error analysis reveals two dominant failure modes: Dialect Particle Blindness (average detection rate 2.9%) and Systematic Phonetic Mismatch (89 vowel confusions), indicating that current models treat dialectal features as noise. These findings underscore the urgent need for dialect-aware ASR adaptation and dataset development, providing a foundation for inclusive speech technology across Indonesia's linguistic diversity.

Related SDGs Targets

Full Author List

Ramadan, Syahril; Indra, Dolly; Azis, Huzain, Universitas Muslim Indonesia

Abstract

Sistem informasi kependudukan merupakan salah satu faktor utama dalam pemerintahan dan pembangunan kependudukan yang diarahakan pada pemenuhan hak dari setiap warga negara dibidang pelayanan data kependudukan. Desa Kaduaja salah satu bagian dari Desa di Kecamatan Gandangbatu Sillanan Kabupaten Tana Toraja, pengelolaan data pada Kantor Desa Kaduaja dilakukan secara manual mengakibatkan dokumen-dokumen tersebut disusun dengan tidak teratur dan tersimpan pada arsip yang terpisah sehingga pihak pemerintah Desa Kaduaja, untuk menyelesaikan permasalahan tersebut dibangunlah sebuah sistem informasi data kependudukan. penelitian ini menghasikan sistem informasi data kependudukan Desa Kaduaja Kecamatan Gandangbatu sillanan berbasis web yang dapat mempermudah pemerintah desa mengolah data kependudukan. metode yang digunakan dalam perancangan aplikasi yaitu metode waterfall yaitu model dimana tiap tahapannya dikerjakan secara berurutan dari atas ke bawah. Pengujian yang dilakukan berdasarkan blackbox testing mendapatkan nilai penggunaan secara keseluruhan dalam tampilan interface maupun fungsionalitas aplikasi yaitu 85% dari 24 responden.

Related SDGs Targets

Full Author List

Purnawansyah; Ramadhan, Ahmad Mufli; Darwis, Herdianti; Lahuddin, Harlinda; Azis, Huzain; Tenripada, Andi Ulfa, Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia

Abstract

Autism Spectrum Disorder (ASD) is fundamentally characterized by impairments in social communication and interaction, as outlined in the DSM-5. Leveraging speech utterances as a non-invasive modality, this study explores binary classification of ASD versus non-ASD based on transcribed utterances. We introduce the Makassar Autism Corpus, a newly constructed small-scale corpus containing utterances in Bahasa Indonesia with a regional Makassar dialect. To enrich the textual data, a set of linguistic features was extracted from each utterance, capturing structural and lexical attributes relevant to ASD indicators. Two classification models were implemented and evaluated: a traditional machine learning approach using Support Vector Machine (SVM) and a deep learning model based on the pretrained Transformer IndoBERT. Experimental results indicate that both models demonstrate the capacity to differentiate ASD from non-ASD utterances, with SVM showing greater stability on low-resource data. This research marks an early effort in applying computational linguistics for ASD classification in Indonesian, highlighting the viability of utterance-based analysis as an objective screening aid aligned with clinical diagnostic frameworks. Despite dataset “limitations, the findings underscore the potential of integrating linguistic feature analysis and natural language processing (NLP) to support early ASD identification in underrepresented language contexts.

Related SDGs Targets

Full Author List

Rusman, Muh; Satra, Ramdan; Azis, Huzain, Universitas Muslim Indonesia

Abstract

The use of the Telegram application as a communication medium allows the exchange of image files that have the potential to become digital evidence, but in some cases the perpetrator deletes the files to eliminate digital traces. This study aims to analyze and prove the ability of the digital forensic process in finding and recovering deleted JPG image files on the Android-based Telegram application so that they can be used as valid digital evidence. The research method refers to the NIST SP 800-86 standard which includes the stages of collection, examination, analysis, and reporting, with the data acquisition process using the logical acquisition method through the Android Debug Bridge (ADB) without root access as well as analysis using FTK Imager and verification of data integrity through SHA-256 hash values. The results of the study show that JPG image files that have been deleted from the Telegram application can still be found and recovered from the internal storage media of Android devices in intact condition, have relevant metadata, and consistent hash values, so they are declared valid as digital evidence

Related SDGs Targets

Full Author List

Azis, Huzain; Ahmad, Randika Sakha Ramadani Putra; Syahriawan, Irawati, Universitas Muslim Indonesia

Abstract

This study presents the implementation of an AI agent chatbot to support journal information services on the Open Journal Systems (OJS) platform and the Telegram messaging application using a dynamic knowledge base sourced from Google Drive. The chatbot provides automated responses to user inquiries related to journal scope, publication fees, submission procedures, and review timelines, while allowing journal administrators to update information content without modifying the system. Functional testing results indicate that the chatbot delivers accurate and consistent information with acceptable response times across both platforms. The implementation demonstrates that integrating an AI agent chatbot with a dynamic knowledge base can enhance information accessibility, reduce administrative workload, and improve service efficiency in academic journal management

Related SDGs Targets

Full Author List

Maharani.S, Putri; Satra, Ramdan, Universitas Muslim Indonesia; Azis, Huzain, Universiti Kuala Lumpur

Abstract

Distributed Denial of Service (DDoS) attacks pose a serious risk to network reliability, particularly within Software-Defined Networking (SDN) architectures that rely on centralized control. This research analyzes the effectiveness of Gaussian Naïve Bayes (GNB) and Support Vector Machine (SVM) with a Radial Basis Function (RBF) kernel for identifying DDoS attacks using an SDN traffic dataset consisting of 104,345 flow records. Three different feature-handling strategies are explored: using the complete feature set without reduction, applying feature selection through SelectKBest, and performing dimensionality reduction with Principal Component Analysis (PCA). Model validation is carried out using Stratified K-Fold Cross-Validation with K values of 2, 5, and 10. Predictions obtained from each fold are merged into a single aggregated confusion matrix to compute evaluation metrics, including accuracy, precision, recall, and F1-score. Experimental results demonstrate that RBF-SVM without feature reduction delivers the highest detection performance, reaching an accuracy of up to 96.9%, while GNB provides lower accuracy but operates with greater computational efficiency. These findings indicate that an evaluation framework based on aggregated confusion matrices can provide more dependable performance estimates for DDoS detection systems deployed in SDN environments.

Related SDGs Targets

Full Author List

Azis, Huzain; Salim, Yulita, Department of Informatics Engineering, Universitas Muslim Indonesia, Jl. Urip Sumoharjo No.km.5, Makassar, 90231, Indonesia; Puspitasari, Yustisia, Orthodontic Department, Dentistry Faculty, Universitas Muslim Indonesia, Jl. Urip Sumoharjo No.km.5, Makassar, 90231, Indonesia

Abstract

In this work, we explore the use of deep learning models for soft tissue profile classification in cephalometric images, an essential task in orthodontic treatment planning. To increase classification accuracy and to preprocess the data and remedy class imbalance, four deep learning architectures were used: Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Recurrent Neural Networks (RNN), and Generative Adversarial Networks (GAN). RNN performed best, with an accuracy of 59.76%, followed by LSTM, with 51.22%. CNN showed signs of overfitting, with a validation accuracy of 95.61% but a testing accuracy of 47.56%. Although the GAN model was used to augment the data and reduce class imbalance, it did not improve classification performance. We used evaluation metrics like precision, recall, and F1 Score to assess how well the models performed. Future work will involve larger datasets and advanced regularization techniques to improve generalizability and clinical relevance. Future directions should focus on expanding the dataset to enhance model accuracy and robustness. Additionally, exploring advanced techniques is essential to prevent overfitting. Furthermore, implementing more sophisticated data augmentation methods and continuously optimizing architectures are recommended to achieve improved classification outcomes and support clinical applications for orthodontic tasks.

Related SDGs Targets

Full Author List

Asran, Asran; Alwi, Erick Irawadi; Azis, Husain, Universitas Muslim Indonesia;

Abstract

Steganografi berfungsi sebagai teknik anti-forensik yang memungkinkan pelaku kejahatan menyembunyikan informasi dalam pesan lain, hal tersebut dapat menyulitkan penyelidik untuk menemukan bukti sebenarnya atas suatu kejahatan. Oleh karena itu, penyelidik harus mengenali dan memanfaatkan alat yang tepat untuk ekstraksi guna mengakses pesan yang melibatkan steganografi. Penelitian ini bertujuan untuk menganalisis bukti digital dengan menerapkan metode forensik statis berdasarkan kerangka kerja yang disiapkan oleh National Institute of Standards and Technology (NIST), penelitian ini juga berfokus pada proses ekstraksi file yang disusupi menggunakan steganografi, yang dilakukan berdasarkan skenario kejahatan. Alat yang dipakai dalam penelitian ini mencakup FTK Imager, Autopsy, HashMyFiles, USB Write Blocker, WinHex, Stegspy, serta InvisibleSecrets. Analisis proses ekstraksi menunjukkan bahwa, dari 10 file yang diperiksa sesuai dengan skenario kasus, 9 file berhasil diekstraksi dan menghasilkan tingkat keberhasilan sebesar 90%. Sebaliknya, 10% file yang tersisa tidak berisi data atau informasi tersembunyi yang relevan. Dari hasil tersebut dapat disimpulkan bahwa file steganografi yang berhasil diekstrak dapat dijadikan sebagai alat bukti materiil hukum yang sah.

Related SDGs Targets

Full Author List

Marlang, Muh Fatwah Fajriansyah; Darwis, Herdianti; Azis, Huzain; Manga, Abdul Rachman, Universitas Muslim Indonesia;

Abstract

Speech in the Bugis-Makassar dialect poses a major challenge for Automatic Speech Recognition (ASR) systems, particularly in detecting clitic particles that are essential to utterance meaning but do not exist in standard Indonesian. This study investigates the effect of synthetic speech data augmentation using OpenAI's gpt-4o-mini-tts on ASR performance for this dialect, testing four ratios of synthetic-to-real data (0%, 50%, 100%, and 150%) through fine-tuning of the wav2vec2-large-xlsr-indonesian model. Evaluation was conducted using Word Error Rate (WER), Character Error Rate (CER), and Particle Detection Rate (PDR) as a dedicated metric for clitic particle detection. Results show that augmentation at a 50% ratio produced consistent improvements across all metrics compared to baseline, with WER decreasing by 3.47 percentage points and PDR increasing by 3.62 percentage points. Further increases in

Related SDGs Targets

Full Author List

Manga, Abdul Rachman; Suat, Wahyu Kadri Rahmat; Azis, Huzain, Universitas Muslim Indonesia;

Abstract

SQL Injection (SQLi) remains a critical cybersecurity threat because it exploits vulnerabilities in user input validation and can compromise the confidentiality, integrity, and availability of information systems. This study compares Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) architectures for detecting malicious SQL queries under identical experimental conditions. Method A publicly available dataset containing 148,327 malicious and benign SQL query instances was preprocessed through missing-value removal, label encoding, tokenization, sequence transformation, padding, and embedding representation. LSTM and CNN models were evaluated using three train-test split scenarios of 70:30, 80:20, and 90:10. Performance was assessed using accuracy, precision, recall, F1-score, and confusion matrices, with the 80:20 split selected for detailed evaluation. Results and Discussion LSTM consistently achieved higher accuracy across the evaluated splits, ranging from 97.84% to 98.00%. Under the 80:20 configuration, LSTM achieved 97.86% accuracy, 99.34% precision, 96.55% recall, and a 97.92% F1-score, compared with CNN at 97.00%, 97.68%, 96.56%, and 97.12%, respectively. LSTM also reduced false positives from 356 to 99, demonstrating better discrimination between legitimate and malicious queries. Conclusion LSTM provides more reliable SQL Injection detection than CNN by better capturing sequential dependencies within SQL query structures, making it a promising approach for practical cybersecurity systems

Related SDGs Targets

Full Author List

Widiyanti, Nabila; Hasanuddin, Tasrif; Azis, Huzain, Universitas Muslim Indonesia;

Abstract

Managing multiple academic tasks with overlapping deadlines remains challenging for university students, while conventional task-management applications still require substantial manual organization and prioritization. This study develops an LLM-based AI Task Agent that enables conversational academic task management through Telegram and workflow automation. Method The proposed system integrates Telegram as the interaction interface, n8n for workflow orchestration, an LLM-based AI agent for natural-language interpretation and tool selection, Google Sheets for task-data operations, PostgreSQL for conversational memory, and scheduled workflows for automated reminders. The system supports Create, Read, Update, and Delete operations, contextual priority recommendations based on deadline, urgency, and lecturer strictness, and proactive reminders. Functional performance and response time were evaluated across the primary system functions. Results and Discussion Create, Read, Update, and Delete operations achieved 100% functional accuracy, while priority recommendation and automated reminder functions achieved 95%. Recorded processing times ranged from 3.1 to 3.5 seconds, with an average of approximately 3.32 seconds. The results demonstrate that separating LLM-based interpretation from predefined external tool execution enables reliable conversational task management while maintaining controlled data operations. Conclusion The proposed LLM-based AI Task Agent demonstrates the feasibility of integrating conversational interaction, executable task-management functions, contextua

Related SDGs Targets