Claim Missing Document
Check
Articles

Found 9 Documents
Search

EKSTRAKSI FITUR BERBASIS AVERAGE FACE UNTUK PENGENALAN EKSPRESI WAJAH Jarot Dwi Prasetyo; Zaehol Fatah; Taufik Saleh
Jurnal Ilmiah Informatika Vol. 2 No. 2 (2017): Jurnal Imliah Informatika
Publisher : Department of Science and Technology Ibrahimy University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35316/jimi.v2i2.464

Abstract

In recent years it appears interest in the interaction between humans and computers. Facial expressions play a fundamental role in social interaction with other humans. In two human communications is only 7% of communication due to language linguistic message, 38% due to paralanguage, while 55% through facial expressions. Therefore, to facilitate human machine interface more friendly on multimedia products, the facial expression recognition on interface very helpful in interacting comfort. One of the steps that affect the facial expression recognition is the accuracy in facial feature extraction. Several approaches to facial expression recognition in its extraction does not consider the dimensions of the data as input features of machine learning Through this research proposes a wavelet algorithm used to reduce the dimension of data features. Data features are then classified using SVM-multiclass machine learning to determine the difference of six facial expressions are anger, hatred, fear of happy, sad, and surprised Jaffe found in the database. Generating classification obtained 81.42% of the 208 sample data.
Optimalisasi Sistem Informasi Untuk Efektifitas Program Kerja Pengawasan Tahunan (PKPT) Inspektorat Kabupaten Bondowoso Lukman Fakih Lidimilah; Jarot Dwi Prasetyo; Ahmad Homaidi
Jurnal Ilmiah Informatika Vol. 5 No. 2 (2020): Jurnal Ilmiah Informatika
Publisher : Department of Science and Technology Ibrahimy University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35316/jimi.v5i2.936

Abstract

The preparation of an annual supervisory work program (PKPT) requires a long process in order to get good and accurate results, because the process is done semi-computerized, which still involves a lot of staff in the Inspectorate. This makes the PKPT formulation process less effective and efficient. So that the supervisory work program that should have been able to be carried out quickly became constrained because of the long process in its preparation. Therefore, it is deemed necessary to design an application that can help facilitate the preparation of an annual monitoring work program with the waterfall method, making it easier for the inspectorate to determine audit objects and the implementation of supervision of predetermined objects. Based on the tests carried out, it was found that the application feasibility was 93%. From the results of these tests, this application can help make it easier for the inspectorate to provide an assessment of the level of risk, determine supervision plans and assignments to be more effective and efficient. So that efforts to become Good Governance and Clean Governance can be realized properly.
EFISIENSI PRESENSI DAN PENGGAJIAN KARYAWAN MENGGUNAKAN QR CODE DAN GEOLOKASI Nur Azizah; Ahmad Homaidi; Irma Yunita; Lukman Fakih Lidimilah; Taufik Saleh; Jarot Dwi Prasetyo; Zaehol Fatah
Conference on Innovation and Application of Science and Technology (CIASTECH) CIASTECH 2022 Transisi Global dalam Mencapai SDGs 2030
Publisher : Universitas Widyagama Malang

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

Abstract

Presensi merupakan sesuatu yang penting karena mempunyai pengaruh terhadap gaji karyawan dan produktivitas lembaga. Proses presensi di lembaga saat ini dilakukan dengan menggunakan model konvensional dengan menandatangani lembar presensi, yang kemudian hasil rekapitulasinya digunakan sebagai perhitungan gaji karyawan. Pengisian presensi tersebut kadang dilakukan menjelang akhir bulan, sehingga tingkat validitasnya berkurang, bahkan kesalahan dalam melakukan rekap presensi menjadi penyebab tidak sesuainya gaji karyawan. Hal tersebut juga mengakibatkan revisi berkepanjangan jika ditemukan setelah proses permohonan pencairan keuangan bulanan. Berdasarkan permasalahan tersebut, dirasa sangat penting untuk membangun sistem informasi untuk menangani presensi dan penggajian karyawan agar lebih efektif dan efisien dengan memanfaatkan QR Code dan geolokasi. Metode Waterfall dengan tahapan communication, planning, modelling, construction, dan deployment menjadi pilihan utama dalam penelitian ini. Hasil yang didapatkan dari penelitian ini adalah sebuah sistem aplikasi yang membantu meiningkatkan efisiensi proses presensi dan penentuan gaji karyawan, sehingga dapat mempercepat dan memudahkan dalam proses pengajuan pencairan keuangan. Hasil pengujian menunjukkan fitur yang disediakan dalam sistem sudah mengakomodir kebutuhan pengguna, sesuai dengan aturan yang berlaku dan dinyatakan valid.
SISTEM INFORMASI PENDISTRIBUSIAN PUPUK BERSUBSIDI BERBASIS WEB Sholehatus Zuhriyah; Taufik Saleh; Irma Yunita; Jarot Dwi Prasetyo
JUSTIFY : Jurnal Sistem Informasi Ibrahimy Vol. 2 No. 1 (2023): JUSTIFY : Jurnal Sistem Informasi Ibrahimy
Publisher : Fakultas Sains dan Teknologi, Universitas Ibrahimy

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35316/justify.v2i1.3429

Abstract

Dinas Tanaman Pangan, Hortikultura da Perkebunan Kab. Situbondo merupakan sebuah organisasi perangkat daerah yang menangani urusan pertanian di Kab.Situbondo. Salah satu program kerjanya adalah dengan melayani petani yang ingin mendapatkan bantuan pupuk bersubsidi. Penetapan kebutuhan dan harga sudah ditetapkan oleh keputusan menteri. Hal ini dilakukan untuk meningkatkan efektivitas penyal pupuk bersubsidi. Pengolahan data pada pendistribusian pupuk bersubsidi masih menggunakan aplikasi Microsoft Word & Microsoft Excel sehingga dapat menimbulkan terjadinya penumpukan data yang menyebabkan terlambatnya pembuatan laporan karena sulitnya mendapatkan informasi dalam waktu cepat. Untuk membantu meminimalisir kekurangan tersebut maka perlu dirancang suatu Sistem Informasi Pendistribusian Pupuk Bersubsidi Berbasis Web. Metode yang digunakan dalam sistem ini adalah dengan menggunakan metode waterfall dengan melalui 5 tahapan diantaranya analisis, desain, pengkodean, pengujian dan supporting. Bahasa pemrogrman yang digunakan pada sistem ini adalah menggunakan bahasa pemrograman Hypertext Prepocesor dan database MySQL. Sistem yang dibangun dapat memberikan informasi dengan cepat, tepat & akurat serta mempercepat pembuatan laporan jika laporan diperlukan bagi pihak-pihak yang membutuhkan.
Classification of Anaemia Status Using The K-Nearest Neighbor Algorithm Zulfa Faradila; Ahmad Homaidi; Jarot Dwi Prasetyo
G-Tech: Jurnal Teknologi Terapan Vol 9 No 1 (2025): G-Tech, Vol. 9 No. 1 January 2025
Publisher : Universitas Islam Raden Rahmat, Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70609/gtech.v9i1.6377

Abstract

Early detection and accurate diagnosis of anemia are crucial for public health management, with conventional methods like complete blood count often being costly and unavailable in remote areas. The use of machine learning techniques, specifically the k-nearest neighbor (KNN) algorithm, shows promise in classifying medical conditions including anemia with competitive accuracy compared to traditional methods. The implementation of KNN not only offers accuracy but also time and cost efficiency, providing reliable results quickly for medical professionals in the field. The algorithm's application involves determining the appropriate k-value for optimal accuracy, calculating distances using Euclidean distance, and voting for class prediction based on nearest neighbors. The analysis showcases the model's efficiency in predicting anemia status with an accuracy of 94.72% and promising precision and recall rates.
Mental Health Diagnosis (Chronic Fatigue Syndrome and Depression) using Decision Tree Algorithm Ach. Zubairi; Ahmad Homaidi; Irma Yunita; Jarot Dwi Prasetyo; Hermanto Hermanto
G-Tech: Jurnal Teknologi Terapan Vol 9 No 3 (2025): G-Tech, Vol. 9 No. 3 July 2025
Publisher : Universitas Islam Raden Rahmat, Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70609/g-tech.v9i3.7595

Abstract

Mental health is an important aspect that affects an individual's life, impacting productivity, social relationships and overall quality of life. The World Health Organization (WHO) states that one in four people worldwide will face mental health challenges. With the increasing incidence of conditions such as depression and Chronic Fatigue Syndrome (CFS), effective detection and intervention methods are urgently needed. Data mining, specifically using Decision Tree algorithms, presents a promising approach to address this challenge. This study utilizes a quantitative methodology to classify depression and CFS patients using a public dataset. The data collection from Kaggle included variables such as demographics and clinical evaluations, consisting of 1,000 records and 15 predictive attributes. Data preprocessing addressed noise, specifically missing values, to ensure model accuracy above 80%. A Decision Tree was implemented, displaying the interpretability of the method by partitioning the data based on the selected attributes. Evaluation metrics such as accuracy, precision, recall, and F1 score showed accuracy of 99% and precision and recall of 100%. The results emphasize the potential of the Decision Tree in distinguishing between depression and CFS, enabling early intervention through accurate patient identification. This study advocates the integration of such machine learning models into clinical practice to improve mental health diagnostics and management, by addressing an important aspect of public health.
PERBANDINGAN ALGORITMA K-NEAREST NEIGHBOUR DAN NAÏVE BAYES UNTUK MENDETEKSI PENIPUAN KARTU KREDIT Fauzan Firdaus; Ahmad Homaidi; Jarot Dwi Prasetyo; Hermanto Hermanto; Ach. Zubairi; Lukman Fakih Lidimilah
Jurnal Ilmiah Informatika Vol. 10 No. 2 (2025): Jurnal Ilmiah Informatika
Publisher : Department of Science and Technology Ibrahimy University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35316/.v10i2.8992

Abstract

Credit card fraud is a serious problem in the financial industry that continues to increase with the development of digital transaction technology. This study aims to compare the performance of the K-Nearest Neighbour (KNN) and Naive Bayes algorithms in detecting credit card fraud by considering various evaluation metrics evaluation metrics, including not only accuracy but also precision, recall, and F1-score. The dataset used was sourced from Kaggle, comprising a total of 10,000 transaction records, which included financial transaction attributes and user behaviour. The research process included data pre-processing, attribute selection, data normalisation, and the application of both algorithms using RapidMiner software. The test results showed that the KNN algorithm produced an accuracy of 98.43%, a precision of 98.53%, and a recall of 99.90%, while Naive Bayes obtained an accuracy 98.20% accuracy, 99.69% precision, and 98.48% recall. Although KNN showed slightly superior performance in detecting fraudulent transactions, the T-Test statistical test showed that the difference in performance between the two algorithms was not statistically significant. KNN has an advantage in recognising complex patterns, but requires greater computational time, while Naive Bayes is more efficient in terms of speed. This study concludes that the selection of a fraud detection algorithm needs to consider the trade-off between accuracy and computational efficiency according to system requirements.
Employee Presence and Payroll Information System Using Quick Response Code and Geolocation Ahmad Homaidi; Lukman Fakih Lidimilah; Jarot Dwi Prasetyo; Nur Azizah
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 23 No. 1 (2023)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v23i1.3093

Abstract

The presence in the institution still uses the conventional method of affixing a signature on the attendance sheet. Presences that have been carried out so far are felt to be less effective and efficient because sometimes attendance is filled towards the end of the month, which causes the validity of attendance data to be questioned. Even errors are often found in the recording, which causes the nominal to be inappropriate and must be revised. This research aimed to design an information system using the Quick Response Code to increase the effectiveness of employee attendance and payroll, supported by geolocation, to make it more efficient. The method used in this research was the waterfall method, using the stages of communication, planning, modeling, construction, and deployment. This research produced an information system that could make it easier for employees to attend, speed up determining employee salaries and filing financial disbursements, and increase employee presence and salary validity. The test results showed that 90% said they were satisfied with the performance of the system being built.
Model Klasifikasi Hybrid Berbasis PSO-KNN untuk Akurasi Diagnosis Penyakit Hepatitis Sunardi Sunardi; Jarot Dwi Prasetyo; Hermanto Hermanto; Ach. Zubairi; Ahmad Homaidi; Irma Yunita; Lukman Fakih Lidimilah
JUSTIFY : Jurnal Sistem Informasi Ibrahimy Vol. 5 No. 1 (2026): JUSTIFY : Jurnal Sistem Informasi Ibrahimy
Publisher : Fakultas Sains dan Teknologi, Universitas Ibrahimy

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35316/justify.v5i1.10322

Abstract

Diagnosis dini penyakit hepatitis masih menghadapi tantangan akibat kompleksitas data medis, tingginya variasi hasil laboratorium, serta keberadaan atribut yang tidak relevan sehingga menurunkan kinerja algoritma K-Nearest Neighbor (K-NN). Penelitian ini bertujuan mengembangkan model hibrida Particle Swarm Optimization–K-Nearest Neighbor (PSO-KNN) untuk mengoptimalkan bobot fitur dan parameter K-NN secara simultan guna meningkatkan akurasi klasifikasi penyakit hepatitis. Metode yang digunakan meliputi preprocessing dataset HCV, normalisasi Min-Max, pembagian data latih dan uji sebesar 80:20, serta optimasi menggunakan PSO dengan 20 partikel dan 30 iterasi. Evaluasi dilakukan menggunakan metrik accuracy, precision, recall, dan F1-score. Hasil penelitian menunjukkan bahwa PSO-KNN meningkatkan akurasi dari 95,1% menjadi 96,7%, presisi dari 88,9% menjadi 100%, recall dari 61,5% menjadi 69,2%, dan F1-score dari 72,7% menjadi 81,8%. Model juga berhasil menghilangkan false positive dan menurunkan false negative. Bobot fitur menunjukkan bahwa ALT, AST, GGT, dan Bilirubin merupakan biomarker yang paling berpengaruh dalam klasifikasi. Dengan demikian, model PSO-KNN terbukti mampu meningkatkan performa klasifikasi, efisiensi komputasi, serta interpretabilitas diagnosis hepatitis, sehingga berpotensi mendukung pengambilan keputusan klinis secara lebih akurat.