Article Per Year (5 Year)
p-Index From 2021 - 2026
2.127
P-Index
This Author published in this journals
All Journal International Journal of Electrical and Computer Engineering IAES International Journal of Artificial Intelligence (IJ-AI) Jurnal Ilmiah Teknik Elektro Komputer dan Informatika (JITEKI) Jurnal TIMES Sinkron : Jurnal dan Penelitian Teknik Informatika JURNAL MEDIA INFORMATIKA BUDIDARMA Building of Informatics, Technology and Science Jurasik (Jurnal Riset Sistem Informasi dan Teknik Informatika) EXPLORER Bulletin of Data Science Bulletin of Informatics and Data Science
Alkhairi, Putrama
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Engineering

Published : 13 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 13 Documents
Search

 1   2 
Optimasi LSTM Mengurangi Overfitting untuk Klasifikasi Teks Menggunakan Kumpulan Data Ulasan Film Kaggle IMDB Alkhairi, Putrama; Windarto, Agus Perdana; Efendi, Muhamad Masjun
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5850

Abstract

This study aims to develop and optimize a Long Short-Term Memory (LSTM) model to reduce overfitting in text classification using the Kaggle IMDB movie review dataset. Overfitting is a common problem in machine learning that causes the model to overfit to the training data, thus degrading its performance on the test data. In this study, various optimization techniques such as regularization, dropout, and careful training methods are applied to improve the generalization of the LSTM model. This study shows that overfitting reduction techniques, such as dropout and the use of the RMSProp optimizer, significantly improve the performance of the Long Short-Term Memory (LSTM) model in IMDB movie review text classification. The optimized LSTM model achieves an accuracy of 83.45%, an increase of 2.07% compared to the standard model which has an accuracy of 81.38%. The precision of the optimized model increases to 89.65%, compared to 84.46% in the standard model, although the recall is slightly lower (75.69% compared to 76.91%). The F1-score of the optimized model is also higher, which is 82.07% compared to 80.53% in the standard model. The experimental results show that the techniques successfully improve the accuracy and reliability of the text classification model, with better performance on the test data. This research makes a significant contribution to understanding and overfitting in deep learning models in the context of natural language processing, and offers insights into best practices in applying LSTM models to text classification.
Effect Effect of Gradient Descent With Momentum Backpropagation Training Function in Detecting Alphabet Letters Alkhairi, Putrama; Batubara, Ela Roza; Rosnelly, Rika; Wanayaumini, W; Tambunan, Heru Satria
Sinkron : jurnal dan penelitian teknik informatika Vol. 7 No. 1 (2023): Articles Research Volume 7 Issue 1, 2023
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v8i1.12183

Abstract

The research uses the Momentum Backpropagation Neural Network method to recognize characters from a letter image. But before that, the letter image will be converted into a binary image. The binary image is then segmented to isolate the characters to be recognized. Finally, the dimension of the segmented image will be reduced using Haar Wavelet. One of the weaknesses of computer systems compared to humans is recognizing character patterns if not using supporting methods. Artificial Neural Network (ANN) is a method or concept that takes the human nervous system. In ANN, there are several methods used to train computers that are made, training is used to increase the accuracy or ability of computers to recognize patterns. One of the ANN algorithms used to train and detect an image is backpropagation. With the Artificial Neural Network (ANN) method, the algorithm can produce a system that can recognize the character pattern of handwritten letters well which can make it easier for humans to recognize patterns from letters that are difficult to read due to various error factors seen by humans. The results of the testing process using the Backpropagation algorithm reached 100% with a total of 90 trained data. The test results of the test data reached 100% of the 90 test data.
Bone fracture classification using convolutional neural network architecture for high-accuracy image classification Solikhun, Solikhun; Windarto, Agus Perdana; Alkhairi, Putrama
International Journal of Electrical and Computer Engineering (IJECE) Vol 14, No 6: December 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijece.v14i6.pp6466-6477

Abstract

This research introduces an innovative method for fracture classification using convolutional neural networks (CNN) for high-accuracy image classification. The study addresses the need to improve the subjectivity and limited accuracy of traditional methods. By harnessing the capability of CNNs to autonomously extract hierarchical features from medical images, this research surpasses the limitations of manual interpretation and existing automated systems. The goal is to create a robust CNN-based methodology for precise and reliable fracture classification, potentially revolutionizing current diagnostic practices. The dataset for this research is sourced from Kaggle's public medical image repository, ensuring a diverse range of fracture images. This study highlights CNNs' potential to significantly enhance diagnostic precision, leading to more effective treatments and improved patient care in orthopedics. The novelty lies in the unique application of CNN architecture for fracture classification, an area not extensively explored before. Testing results show a significant improvement in classification accuracy, with the proposed model achieving an accuracy rate of 0.9922 compared to ResNet50's 0.9844. The research suggests that adopting CNN-based systems in medical practice can enhance diagnostic accuracy, optimize treatment plans, and improve patient outcomes.
Deep Learning to Extract Animal Images With the U-Net Model on the Use of Pet Images Windarto, Agus Perdana; Rahadjeng, Indra Riyana; Siregar, Muhammad Noor Hasan; Alkhairi, Putrama
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7199

Abstract

This article explores the innovative application of deep learning techniques, specifically the U-Net model, in the realm of computer vision, focusing on the extraction of animal images from diverse pet datasets. As the digital landscape becomes increasingly saturated with pet imagery, the need for precise and efficient image extraction methods becomes paramount. The study delves into the challenges posed by varying animal poses and backgrounds, presenting a comprehensive analysis of the U-Net model's adaptability in handling these complexities. Through rigorous experimentation, this research refines existing methodologies, enhancing the accuracy of animal image extraction. The findings not only contribute to advancing the field of computer vision but also hold significant implications for wildlife monitoring, veterinary diagnostics, and the broader domain of image processing.
A revolutionary convolutional neural network architecture for more accurate lung cancer classification Muliadi, Muliadi; Windarto, Agus Perdana; Solikhun, Solikhun; Alkhairi, Putrama
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 1: February 2025
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v14.i1.pp516-526

Abstract

This research aimed to investigate a breakthrough in convolutional neural network (CNN) architecture with the potential to revolutionize lung cancer classification. The proposed method is a comparative optimization model of ResNet architecture, with accuracy rate of 99.68% in identifying and categorizing lung cancer types. The results showed that the use of innovative methods in CNN architecture, such as multi-dimensional convolutional layers and the integration of specific lung cancer features, effectively provided highly accurate and reliable outcomes. These showed a positive impact on the development of medical diagnostic technology, offering promising potential to enhance prognosis and response to treatment for lung cancer patients. With high accuracy rate, this breakthrough presents opportunities for further advancements in lung cancer management through artificial intelligence-based methods.
Penggunaan Jaringan Saraf Tiruan untuk Memperkirakan Tenaga Kerja Berdasarkan Kategori Industri Ariani, Dhini; Saragih, Farah Yusni; Asyifah, Hazha Hikmah; Nasution, Alisa Putri Amanda; Alkhairi, Putrama
Bulletin of Data Science Vol 4 No 1 (2024): October 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletinds.v4i1.6433

Abstract

Industrial growth can affect labor mobility both geographically and in terms of professional qualifications, large industries have a strategic role as creators of added value and important job providers in the Region. As an important part of industrial production, it cannot be separated from the demand for labor, but if viewed macro, it can be concluded that the quality of work determines or greatly influences the results of labor productivity itself. The industrial sector plays a significant role in economic growth, because it absorbs labor. Labor growth is much greater than the availability of jobs, thus causing other new problems, namely high unemployment. This study uses the Backpropagation Method to classify special patterns, which reduces the error rate by adjusting the weight based on the difference between output and the desired target. The results of this study are predictions of the level of truth of the Number of Large and Medium Industrial Workers according to Industry Group. Using 5 models, namely 10-10-1, 10-45-1, 10-45-10-1, 10-75-10-1, and 10-100-75-1. From 5 architecture models, 1 best model is produced, namely the 10-75-10-1 model with an accuracy rate of 70% and the smallest epoch with a total of 383.
Optimasi Fungsi Aktivasi pada Artificial Neural Network untuk Prediksi Gagal Jantung Secara Akurat Raharjo, Mokhamad Ramdhani; Indra Riyana Rahadjeng; Siregar, Muhammad Noor Hasan; Alkhairi, Putrama
Explorer Vol 5 No 1 (2025): January 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/explorer.v5i1.1840

Abstract

Heart failure is one of the major health problems that can be fatal if not diagnosed properly and quickly. Therefore, early prediction using artificial intelligence models, especially Artificial Neural Network (ANN), is needed to improve the accuracy in detecting heart failure. This study aims to optimize the activation function in ANN to predict heart failure accurately. Several optimization algorithms tested, namely Adam, RMSprop, SGD, Adagrad, and Adadelta, were used to evaluate model performance in terms of accuracy, precision, recall, and F1-score. The results showed that the Adam optimization algorithm provided the best performance with an accuracy of 86.74%, precision of 75.12%, recall of 66.67%, and F1-score of 70.64%. Meanwhile, other algorithms such as RMSprop, SGD, Adagrad, and Adadelta showed lower performance, with some metrics reaching 0%. This study shows that proper activation function optimization in ANN is very important to improve the model's ability to predict heart failure with a high level of accuracy.
Penerapan Data Mining Untuk Menganalisis Kepuasan Pegawai Terhadap Pelayanan Bidang SDM dengan Algoritma C4.5 Alkhairi, Putrama; Situmorang, Zakarias
Jurasik (Jurnal Riset Sistem Informasi dan Teknik Informatika) Vol 7, No 1 (2022): Edisi Februari
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/jurasik.v7i1.414

Abstract

Employee satisfaction includes the difference between the level of importance and perceived performance or results, and is an alternative evaluation that exceeds employee expectations. There are 5 dimensions to measure service quality based on expectations and perceived performance by employees, namely career development, leadership in HR, policy and law enforcement, building a work atmosphere and providing salaries and rewards. Five dimensions are very influential in the progress of STIKOM Tunas Bangsa, using data mining methods can be found important trends for campuses. Employee satisfaction assessment is based on a questionnaire filled out by the employee. The results of the questionnaire were processed using the c4.5 algorithm. The c4.5 algorithm is a classification method and produces a decision tree. C4.5 turns large facts into decision trees that represent rules. Rules are easy to understand in natural language. Based on the results of the research that has been done, the use of the C4.5 algorithm can help the campus in improving services according to the results of the questionnaire. The results of the calculation, there are two variables satisfied employee questionnaire. Meanwhile, the employee questionnaire was not satisfied with the three variables. The highest gain value is the variable to build a work atmosphere with a value of 0.20619372. The indicator of the variable of building a work atmosphere that has the highest entropy value is a fairly good indicator with a value of 1. The total of questionnaires filled in are 65 questionnaires, 44 people stated they were satisfied and only 21 people said they were not satisfied.
Optimasi JST Backpropagation dengan Adaptive Learning Rate Dalam Memprediksi Hasil Panen Padi Prihandoko, P; Alkhairi, Putrama
Jurasik (Jurnal Riset Sistem Informasi dan Teknik Informatika) Vol 10, No 1 (2025): Edisi Februari
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/jurasik.v10i1.887

Abstract

Artificial Neural Networks (ANN) with the Backpropagation algorithm have been widely applied across various domains, including data prediction tasks. However, one of the primary challenges in implementing Backpropagation is the selection of an optimal learning rate. A learning rate that is too high can lead to unstable convergence, while one that is too low can significantly slow down the training process. To address this issue, this study proposes an optimization of Backpropagation using an Adaptive Learning Rate through the implementation of the Adam optimizer. The objective of this research is to analyze the performance comparison between Standard Backpropagation and Backpropagation with the Adam optimizer in predicting rice harvest yields based on rainfall, temperature, and humidity variables. The dataset consists of 100 synthetic samples generated based on a normal distribution to resemble real-world data. The results show that the use of the Adam optimizer improves the performance of the ANN model compared to the Standard Backpropagation method. Model accuracy increased from 92.04% to 92.99%, while the values of loss, Mean Squared Error (MSE), and Root Mean Squared Error (RMSE) decreased significantly, indicating that the model optimized with Adam is more stable and yields lower prediction errors. Therefore, Adaptive Learning Rate optimization using the Adam optimizer is proven to be effective in enhancing both the accuracy and efficiency of ANN in data prediction tasks.
PENGENALAN POLA KEMAMPUAN PELANGGAN DALAM MEMBAYAR AIR PDAM MENGGUNAKAN ALGORITMA NAÏVE BAYES Ilmi R.H. Zer, P.P.P.A.N.W. Fikrul; Batubara, Ela Roza; Alkhairi, Putrama; Tambunan, Fazli Nugraha; Rosnelly, Rika
Jurnal TIMES Vol 10 No 2 (2021): Jurnal TIMES
Publisher : STMIK TIME

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (828.686 KB) | DOI: 10.51351/jtm.10.2.2021656

Abstract

Dengan meningkatnya jumlah MBR (Masyarakat Berpenghasilan Rendah) yang masuk setiap tahunnya dimasing-masing wilayah di Pematansgsiantar, pihak PDAM Tirta Lihou berencana mencari alternatif solusi dalam menangani permasalahan kemampuan pelanggan dalam membayar tagihan air sehingga biaya opersional tetap bisa berjalan baik dan produksi dapat memenuhi kebutuhan masyarakat. Dalam menentukan alternatif untuk menentukan kemampauan masyarakat dalam membayar tagiahan air digunakan metode datamining. Dengan menggunakan teknik datamining khususnya klasifikasi menggunakan algoritma Naive Bayes dapat dilakukan prediksi terhadap kemampauan pelanggan dalam membayar tagihan air bersih berdasarkan data yang ada. Naive bayes adalah teknik prediksi probabilistik sederhana yang berdasarkan pada teorema Bayes dengan asumsi independensi (ketidak tergantungan) yang kuat. Berdasarkan hasil dari perhitungan menggunakan algoritma Naive Bayes, diperoleh hasil klasifikasi dari 30 alternatif yang digunakan, dimana terdapat 11 kelas mampu membayar tagihan dan 19 Tidak Mampu dengan total Accuracy yang diperoleh sebesar 70%. Dari hasil yang diperoleh,diharapkan penelitian ini dapat membantu pihak PDAM Tirta Lihou dalam menentukan lokasi yang layak dilakukan penaybungan sumber air untuk pelanggan yang memiliki prosfek baik dengan kemampuan untuk membayar tagihan air, sehingga dapat meminimalisir kerugian PDAM dan dapat memenuhi kebutuhan masyarakat. Penelitian ini juga diharapkan dapat menjadi referensi bagi peneliti selanjutnya yang berkaitan dengan pengguna algoritma yang digunakan.
Co-Authors Agus Perdana Windarto Ahmad Zamsuri, Ahmad Ariani, Dhini Asyifah, Hazha Hikmah Batubara, Ela Roza Efendi, Muhamad Masjun Heru Satria Tambunan, Heru Satria Ilmi R.H. Zer, P.P.P.A.N.W. Fikrul Indra Riyana Rahadjeng Indra Riyana Rahadjeng Irawan Irawan Muhammad Noor Hasan Siregar Muliadi Nasution, Alisa Putri Amanda Nuroini, Indi Pardede, Akim Manaor Hara Parinduri, Syawaluddin Kadafi Poningsih Poningsih, Poningsih Prihandoko, P Qurniawan, Hendry Raharjo, Mokhamad Ramdhani Rika Rosnelly, Rika Saragih, Farah Yusni Solikhun Solikhun, Solikhun Tambunan, Fazli Nugraha Wanayaumini, W Zakarias Situmorang