Claim Missing Document
Check
Articles

Found 21 Documents
Search

Pengembangan Aplikasi Sistem Pendukung Keputusan Rekrutmen Perawat Menggunakan Metode Weighted Product Dedy Armiady
Jurnal Tika Vol 8 No 1 (2023): Jurnal Teknik Informatika Aceh
Publisher : Fakultas Ilmu Komputer Universitas Almuslim Bireuen - Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51179/tika.v8i1.1857

Abstract

Work that requires high accuracy and must be done repeatedly is work that must be completed using a computer. This work will certainly be very difficult to complete if you only rely on human power which has weaknesses in terms of memory and speed. Currently, many agencies and companies are aggressively implementing computerized technology to support good management, one of which is a hospital. In addition to patient data management and other important data, cases of nurse recruitment are also things that need to be considered to be resolved by the computer. This study aims to develop a web-based decision support system using a weighted product model to calculate alternative rankings. The results obtained are that the weighted product method can be used to calculate the ranking of prospective nurses through 12 specified criteria, where from several applicants, 3 prospective nurses with the highest rank are taken. The decision support system application in this study was developed using the PHP and MySQL programming languages with the CodeIgniter framework
Pengembangan Sistem Informasi Pengajuan Judul Skripsi Menggunakan Algoritma Winnowing Dedy Armiady
Jurnal Tika Vol 8 No 2 (2023): Jurnal Teknik Informatika Aceh
Publisher : Fakultas Ilmu Komputer Universitas Almuslim Bireuen - Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51179/tika.v8i2.2066

Abstract

Administrative implementation in higher education institutions is currently increasingly directed to completely transform into digital-based administration by utilizing computer-based technology. It is intended that all forms of administrative services can be neatly arranged and easy to access. One case that has also become a concern for development is the case of submitting a student thesis title. Currently, many systems are being developed just for CRUD (Create, Read, Delete and Update) needs. Where in the old system, there was no similarity detection feature for proposed titles with existing titles. This study aims to build a system for submitting thesis titles for students of the Faculty of Computer Science, Almuslim University which has a title check similarity feature. In this study, the winnowing algorithm is used as a mathematical calculation model to calculate the proportion of document similarities using the document fingerprint method. The developed system is a web-based information system using the PHP and MySQL programming languages. As for the results obtained, the winnowing algorithm can be applied to a title submission information system that can display the proportion of the proposed title to the existing title
Klasifikasi Kualitas Buah Pisang Berdasarkan Citra Buah Menggunakan Stochastic Gradient Descent Dedy Armiady; Imam Muslem R
KLIK: Kajian Ilmiah Informatika dan Komputer Vol. 4 No. 2 (2023): Oktober 2023
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/klik.v4i2.1243

Abstract

Banana fruit quality is an important factor in meeting consumer demand and maintaining product quality in the supply chain. The development of automatic methods for classifying the quality of bananas is becoming increasingly important as the worldwide consumption of bananas grows. In this study, we propose a classification method for banana fruit quality using the Stochastic Gradient Descent (SGD) algorithm. This study aims to evaluate the performance of SGD in classifying the quality of bananas and to analyze the effect of selecting hyperparameters on the classification results. The dataset collected is a dataset containing pictures of bananas with various levels of ripeness and conditions. This dataset is used to train and test a classification model using SGD. During the experiment, hyperparameter tuning processes such as learning rate, momentum, and batch size were carried out to understand how these parameters affect the performance of SGD in classification. We report the results of evaluating the classification based on accuracy and analyze changes in performance with variations in hyperparameters. The results of this study indicate that SGD has the potential to classify the quality of bananas, where the optimal SGD model obtained a classification accuracy of 99.9%, compared to the standard SGD model which only obtained a classification accuracy of 94.7%.
Analisis Algoritma Logistic Regression dan Support Vector Machine pada Kasus Klasifikasi Citra Hewan Rawa dengan Dataset yang tidak Seimbang Armiady, Dedy
Data Sciences Indonesia (DSI) Vol. 4 No. 1 (2024): Article Research Volume 4 Issue 1, June 2024
Publisher : Yayasan Cita Cendikiawan Al Kharizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/dsi.v4i1.4433

Abstract

Penelitian ini bertujuan untuk mengevaluasi kinerja dua algoritma machine learning, Logistic Regression dan Support Vector Machine (SVM), dalam tugas klasifikasi citra hewan rawa menggunakan dataset yang tidak seimbang. Dataset yang digunakan terdiri dari empat kategori hewan rawa: Buaya, Kodok, Kura-kura, dan Ular, dengan distribusi yang sangat tidak merata. Kelas Kura-kura memiliki jumlah sampel yang jauh lebih banyak dibandingkan dengan kelas lainnya, menciptakan tantangan ketidakseimbangan data yang signifikan. Metode penelitian dimulai dengan mengimpor dataset citra hewan rawa ke dalam tool Orange Data Mining, diikuti oleh proses ekstraksi fitur menggunakan SqueezeNet (local) sebagai embedder. Dua model machine learning, yaitu Logistic Regression dan SVM, kemudian dilatih menggunakan fitur yang diekstraksi. Evaluasi model dilakukan dengan menambahkan widget test and score untuk mengukur metrik performa seperti Area Under the Curve (AUC), akurasi klasifikasi (CA), F1-Score, precision, recall, dan Matthews Correlation Coefficient (MCC). Hasil penelitian menunjukkan bahwa Logistic Regression unggul dalam hampir semua metrik evaluasi dibandingkan SVM. Logistic Regression mencapai nilai AUC sebesar 0.985, akurasi klasifikasi 0.908, F1-Score 0.909, precision 0.909, recall 0.908, dan MCC 0.859. Sebaliknya, SVM mencapai nilai AUC 0.971, akurasi klasifikasi 0.863, F1-Score 0.867, precision 0.877, recall 0.863, dan MCC 0.797. Kesimpulan dari penelitian ini adalah bahwa Logistic Regression merupakan model yang lebih tepat untuk tugas klasifikasi citra hewan rawa dengan dataset yang tidak seimbang. Model ini tidak hanya menunjukkan kinerja yang lebih baik dalam membedakan kelas-kelas citra tetapi juga lebih akurat dan seimbang dalam mengklasifikasikan sampel dari kelas minoritas
Klasifikasi Jenis Jamur Berdasarkan Citra Gambar Menggunakan Algoritma Stochastic Gradient Descent Armiady, Dedy
Data Sciences Indonesia (DSI) Vol. 4 No. 2 (2024): Article Research Volume 4 Issue 2, December 2024
Publisher : Yayasan Cita Cendikiawan Al Kharizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/dsi.v4i2.5014

Abstract

Penelitian ini mengkaji penerapan Stochastic Gradient Descent (SGD) untuk klasifikasi gambar jamur berdasarkan citra, menggunakan dataset tidak seimbang dengan 10 kelas jamur yang berbeda. Algoritma SGD digunakan karena efisiensinya dalam menangani dataset besar serta kemampuan untuk memperbarui parameter secara bertahap guna meminimalkan fungsi loss. Penyesuaian parameter pada SGD, seperti Squared Hinge Loss untuk klasifikasi, Elastic Net sebagai regularisasi, dan optimal learning rate, dilakukan untuk meningkatkan performa model. Hasil penelitian menunjukkan bahwa akurasi model mencapai 62.4%, yang lebih tinggi dibandingkan dengan beberapa algoritma lain seperti CNN, Logistic Regression, SVM, dan Random Forest yang telah diuji pada dataset yang sama. Meskipun demikian, nilai AUC sebesar 0.775 dan F1-score 0.625 menunjukkan bahwa performa model masih belum optimal, terutama dalam menangani ketidakseimbangan data. Penggunaan teknik penyeimbangan data seperti SMOTE direkomendasikan untuk penelitian lebih lanjut guna meningkatkan performa klasifikasi pada kelas yang minoritas. Dengan penyesuaian parameter yang tepat, SGD terbukti mampu bekerja lebih baik dibandingkan model lain dalam konteks klasifikasi citra jamur. Hasil ini memberikan kontribusi signifikan dalam bidang klasifikasi berbasis citra, khususnya pada aplikasi dengan dataset yang tidak seimbang
A PSO-Based CVRPPD Model with Weather and Traffic Constraints for Two-Wheeled Urban Delivery in Indonesia Muslem R, Imam; Armiady, Dedy
Bahasa Indonesia Vol 17 No 05 (2025): Instal : Jurnal Komputer
Publisher : Cattleya Darmaya Fortuna

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54209/jurnalinstall.v17i05.388

Abstract

Delivery of goods using two-wheeled vehicles is a dominant model of logistical distribution in urban areas of Indonesia, yet it faces practical challenges such as limited load capacity, traffic congestion, and adverse weather conditions. This study develops a model of the Capacitated Vehicle Routing Problem with Pickup and Delivery (CVRPPD) based on the Particle Swarm Optimization (PSO) algorithm to generate optimal routes that are both efficient and adaptive to these constraints. The model utilizes customer spatial data, shipment loads (both delivery and pickup), and environmental penalty factors in the form of weather and traffic levels, represented on an ordinal scale. The simulation process was executed over 200 iterations, yielding the best solution at iteration 60 with a best cost value of 2554.84 and a total of 13 vehicle routes, each complying with the maximum load capacity of 20 kg. The results indicate that PSO is capable of generating balanced, efficient, and realistic route allocations by explicitly accounting for the operational challenges faced by couriers in the field. This study contributes to the development of data-driven and AI-based logistics optimization systems that are contextualized for urban environments in developing countries.
Penetapan Klaster Siswa Unggul Dengan Menggunakan Algoritma Roc-Smarter Armiady, Dedy; Muslem R., Imam
Jurnal Tika Vol 7 No 2 (2022): Jurnal Teknik Informatika Aceh
Publisher : Fakultas Ilmu Komputer Universitas Almuslim Bireuen - Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51179/tika.v7i2.1229

Abstract

The development of the IT world today has penetrated into all sectors of human life, one of which is the education sector. However, the implementation that occurs in the field has not been fully maximized, considering that there are still educational institutions that still seem half-hearted in implementing information technology. This happens due to various problems, such as lack of funds for IT development, IT infrastructure that is still minimal, especially for educational institutions located far from urban areas, human resources that are still lacking in skills and knowledge and various factors. Many efforts can be taken to improve this, one of which is as carried out in this study. The focus of this research is to develop an application by applying the ROC-SMARTER method so that it can be used by MTsN 2 Bireuen in determining the superior student cluster. From the implementation of the research using the ADDIE approach (Analyze, Design, Development, Implementation, and Evaluation) it was found that the case of determining the cluster of superior students at MTsN 2 Bireuen could be solved effectively and efficiently where previously it was done using manual and traditional methods
Pengembangan Aplikasi Berbasis Responsive Website Menggunakan Metode Extreme Programming Armiady, Dedy
Jurnal Tika Vol 7 No 2 (2022): Jurnal Teknik Informatika Aceh
Publisher : Fakultas Ilmu Komputer Universitas Almuslim Bireuen - Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51179/tika.v7i2.1266

Abstract

The use of information technology, especially computer network-based applications, is currently increasingly widespread. This is indicated by the number of educational institutions that have begun to apply this technology to solve various problems faced. One of the educational institutions currently committed to implementing information technology is the Almuslim Integrated Islamic Boarding School. This study aims to develop an information system for the registration of prospective new students at the Almuslim Integrated Islamic Boarding School based on a responsive website using the Extreme Programming (XP) method. This method is used to achieve the ultimate goal of system development quickly and effectively. Development Based on the system implemented, this method is able to accommodate the system development process quickly with a focus on the process of writing program code. The results obtained are a responsive website-based information system for registration of prospective new students which was developed in a short time. This result is different from the development of the previous system using the waterfall model
Identifikasi Tingkat Kematangan Buah Naga Merah (Hylocereus Costaricensis) Melalui Pendekatan Artificial Neural Network (Ann) Armiady, Dedy
Jurnal Tika Vol 7 No 3 (2022): Jurnal Teknik Informatika Aceh
Publisher : Fakultas Ilmu Komputer Universitas Almuslim Bireuen - Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51179/tika.v7i3.1576

Abstract

Massive technological developments continue to occur and penetrate into all sectors of the world's people's lives. In Indonesia in particular, various studies need to be carried out to develop various 4.0 technologies in agriculture and apply them to improve the quality and quantity of production. One of the technologies in agriculture that needs to be developed is the identification of fruit maturity, where this needs to be done considering the limitations of the human senses in determining the level of maturity based on the RGB value of the fruit. In this study, an Artificial Neural Network (ANN) approach with the Backpropagation algorithm was used. The dataset used consists of 90 photos of dragon fruit for training data and 15 photos of dragon fruit for data testing. The results obtained are that the ANN model built is able to identify the level of fruit maturity with 100% accuracy based on the dataset used
Pengembangan Aplikasi Sistem Pendukung Keputusan Rekrutmen Perawat Menggunakan Metode Weighted Product Armiady, Dedy
Jurnal Tika Vol 8 No 1 (2023): Jurnal Teknik Informatika Aceh
Publisher : Fakultas Ilmu Komputer Universitas Almuslim Bireuen - Aceh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51179/tika.v8i1.1857

Abstract

Work that requires high accuracy and must be done repeatedly is work that must be completed using a computer. This work will certainly be very difficult to complete if you only rely on human power which has weaknesses in terms of memory and speed. Currently, many agencies and companies are aggressively implementing computerized technology to support good management, one of which is a hospital. In addition to patient data management and other important data, cases of nurse recruitment are also things that need to be considered to be resolved by the computer. This study aims to develop a web-based decision support system using a weighted product model to calculate alternative rankings. The results obtained are that the weighted product method can be used to calculate the ranking of prospective nurses through 12 specified criteria, where from several applicants, 3 prospective nurses with the highest rank are taken. The decision support system application in this study was developed using the PHP and MySQL programming languages with the CodeIgniter framework