Claim Missing Document
Check
Articles

Optimizing LSTM with Grid Search and Regularization Techniques to Enhance Accuracy in Human Activity Recognition Budiarso, Zuly; Listiyono, Hersatoto; Karim, Abdul
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.433

Abstract

This study aims to enhance the accuracy of Long Short-Term Memory (LSTM) models for human activity recognition using the UCI Human Activity Recognition (HAR) dataset. The dataset comprises time-series data from accelerometer and gyroscope sensors on smartphones worn by 30 volunteers as they performed everyday activities such as walking, climbing stairs, descending stairs, sitting, standing, and lying down. Optimization was carried out using Grid Search for hyperparameter tuning and L2 regularization to prevent overfitting. The results show that the optimized LSTM model improved accuracy from 92.33% to 94.50%, precision from 93.12% to 94.61%, recall from 92.33% to 94.50%, and F1-score from 92.32% to 94.51% compared to the standard LSTM model. Despite these improvements, the study encountered several challenges, particularly in tuning hyperparameters, which required significant computational resources and time due to the complexity of the search space. Additionally, balancing regularization to prevent both underfitting and overfitting proved to be a delicate process. Further limitations include the model's performance variability with different sensor placements and potential overfitting to specific activity patterns. However, the implementation of hyperparameter optimization and regularization proved effective in improving the model's ability to recognize human activity patterns from complex sensor data. Therefore, this approach holds significant potential for broader applications in sensor-based human activity recognition systems, though further research is needed to address these limitations and generalize the findings.
Perhitungan Bandwith Dan Spesifikasi Client Server Local Meeting Untuk Mengatasi Lonjakan Trafic Yulianton, Heribertus; Priambodo, Ilyas; Listiyono, Hersatoto
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 7, No 1 (2023): EDISI MARET
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/j-sakti.v7i1.594

Abstract

The Pandemic period had an impact on the meeting system at dr. Kariadi Hospital where internal meetings which are usually held face-to-face are now required through virtual media. To overcome the excessive use of the internet, a local meeting server was created using the Linux Debian Buster Operating System and the OpenMeetings Application System. This study aims to find local meeting server resources that suit the needs to be comfortable to use by users. To achieve the  convenience  and  its application  in  using  the  meeting  server  application  required  resources.  Multiple Linear Regression predicts the amount of bandwidth by using computer resource usage data on the number of application users. By creating a local meeting server and calculating server resources, it helps to expedite and make it easier to carry out meetings that must be held in different places.
Rekayasa Arsitektur Buku Tamu Digital Purwatiningtyas, P; Rani, R.P.S.; Listiyono, Hersatoto; Mulyani, Sri
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 7, No 2 (2023): EDISI SEPTEMBER
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/j-sakti.v7i2.697

Abstract

The guest book is a means that can be used to find out how many guests have come to a place or other uses that require data from visitors to find out about service activities in an institution. Currently there are still many office agencies that use manual or handwritten methods. into the book to record the guest list before entering the office area. Then the book will be stored in the guard post. This is of course considered less effective because it will pose various risks. To anticipate the risks that will occur, a web-based digital guest book is created that can change the use of paper guest books. This guest book was created using the programming language HTML, PHP and the Bootstrap framework and created using the Waterfall method. Guest data will be saved automatically into the database that has been provided and can be processed by the admin. The final result of this research is a digital guest book that can make it easier for officers to store archives, process data, and print data when needed.
Klasifikasi Opini Pengguna Media Sosial Twitter Terhadap JNT Di Indonesia dengan Algoritma Decision Tree Handoko, Widiyanto Tri; Supriyanto, Edy; Purwadi, Dimas Indra; Budiarso, Zuly; Listiyono, Hersatoto
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 6, No 2 (2022): EDISI SEPTEMBER
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/j-sakti.v6i2.490

Abstract

JNT Ekspress is one of the many freight forwarding companies that exist today, where JNT has very wide access so it is very easy to use for the public in shipping goods. With the current network, JNT is able to deliver goods to all provinces in Indonesia. With the large number of users, of course there will be a lot of user opinions that appear, both positive and negative opinions. In order to be able to categorize multiple opinions, a machine learning program is needed that can simplify the process of grouping the opinion. There are many algorithms that can be used to classify opinions, one of them is Decision Tree. Prior to grouping or classification, Tweet data that has been collected needs to be preprocessed first so that the tweet data can be recognized by the system. Based on this research, the Decision Tree algorithm gets an accuracy of 94.12% with a comparison ratio of training data and testing data of 90:10
Pemodelan SPK Persetujuan Ajuan Pinjaman Dengan Mengimplementasikan Metode Smart Dan Kriteria 7P Listiyono, Hersatoto; Purwatiningtyas, P; Sunardi, S; Maskur, Ali; Supriyanto, Edi
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.v10i1.853

Abstract

This research aims to explore the utilization of a Decision Support System (DSS) in the loan application approval process using the SMART method. DSS helps identify, assess, and mitigate various types of risks associated with loans, enhancing the efficiency and productivity of loan processing operations. By automating repetitive tasks, streamlining workflows, and reducing manual intervention, it improves customer satisfaction by providing faster response times. DSS also helps lenders comply with applicable laws and regulations by incorporating compliance checks, regulatory guidelines, and audit trails into the loan processing workflow, thereby reducing the costs and risks associated with manual processes, errors, and inefficiencies in loan processing. Traditional loan assessment methods often require manual processes, leading to delays and inaccuracies. The SMART method (Specific, Measurable, Achievable, Relevant, Time-bound) offers a structured approach to implementing DSS with clear and measurable outcomes. By leveraging DSS, lenders can streamline the loan application process, enhance decision-making accuracy, and improve customer satisfaction. This study will investigate the use of a SMART-based DSS in the loan application workflow and provide insights into best practices for implementing a SMART-based DSS solution in lending institutions. The findings of this research will contribute to the optimization of loan application processes and the improvement of decision-making capabilities in the financial industry, consistently, objectively, and more quickly. The system has the capability to collect, store, and manage data related to loan applications.
DETEKSI PHISHING WEBSITE MENGGUNAKAN SUPPORT VECTOR MACHINE, GRADIENT BOOSTING, DAN NEURAL NETWORKS Indriani, Vanyariska; Listiyono, Hersatoto; Saefurrahman, Saefurrahman
JATI (Jurnal Mahasiswa Teknik Informatika) Vol. 9 No. 2 (2025): JATI Vol. 9 No. 2
Publisher : Institut Teknologi Nasional Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36040/jati.v9i2.13214

Abstract

Di era digital, phishing menjadi ancaman yang signifikan terhadap keamanan data pribadi pengguna internet. Website phishing yang meniru tampilan situs asli bertujuan untuk memperoleh informasi sensitif seperti data pribadi dan keuangan. Penelitian ini bertujuan untuk menguji efektivitas tiga algoritma machine learning, yaitu Support Vector Machine (SVM), Gradient Boosting, dan Neural Networks dalam mendeteksi URL phishing. Dataset PhiUSIIL yang terdiri dari 235.795 sampel data digunakan, yang mencakup URL phishing dan legitim dengan 54 fitur. Metode yang diterapkan meliputi preprocessing data, pembagian data latih dan uji, serta normalisasi menggunakan MinMaxScaler. Ketiga model dievaluasi berdasarkan akurasi, F1-Score, Confusion Matrix, dan waktu komputasi. Hasil penelitian menunjukkan bahwa model Gradient Boosting memperoleh akurasi dan F1-Score 100%, dengan waktu komputasi tercepat hanya 1 menit, menjadikannya pilihan terbaik untuk deteksi phishing. Sementara itu, SVM dan Neural Networks juga menunjukkan hasil yang sangat baik, masing-masing dengan akurasi 99,99%, namun dengan waktu komputasi yang lebih lama. Berdasarkan hasil tersebut, penelitian ini menyimpulkan bahwa Gradient Boosting adalah model yang paling efisien untuk mendeteksi phishing pada dataset yang digunakan.
Pelatihan Digital Marketing Siswa SMKN-01 Tonjong Brebes Untuk Meningkatkan Kapasitas Pemasaran Pada Era Digital Listiyono, Hersatoto; Yunus Anis; Dwi Agus Diartono; Isworo Nugroho
Servis : Jurnal Pengabdian dan Layanan kepada Masyarakat Vol. 2 No. 1 (2023): Desember
Publisher : CV. Nature Creative Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58641/servis.v2i1.60

Abstract

Pelatihan digital marketing adalah suatu pendekatan pendidikan yang krusial dalam era digital saat ini. Siswa Sekolah Menengah Kejuruan (SMK) memiliki potensi besar untuk mengembangkan keterampilan dalam bidang ini. Pelatihan digital marketing memberikan pemahaman yang mendalam tentang strategi pemasaran digital, alat, dan teknik yang dapat siswa terapkan di dunia nyata. Manfaat penting pelatihan digital marketing bagi siswa SMK mencakup peningkatan permintaan pekerjaan di bidang ini, peluang berwirausaha, kemampuan adaptasi terhadap teknologi baru, dan kesempatan untuk berpartisipasi dalam proyek-proyek praktis. Pelatihan ini juga memungkinkan siswa untuk memahami pasar global, menyediakan rujukan akademis, dan mengurangi kesenjangan kompetensi di antara lulusan SMK. Metode pelatihan yang efektif mencakup kelas daring, pembelajaran berbasis proyek, kolaborasi dengan industri, pelatihan dalam kelompok kecil, mentoring, sertifikasi digital marketing, kompetisi, evaluasi, dan klub digital marketing. Semua ini bertujuan untuk mempersiapkan siswa SMK agar siap menghadapi tantangan dan peluang di dunia kerja yang semakin digital.Pentingnya pelatihan digital marketing untuk siswa SMK adalah investasi dalam masa depan pengembangan sumber daya manusia yang kompeten dalam menghadapi dunia digital yang terus berkembang.
Pelatihan Tracer Study Siswa SMK-01 Tonjong Brebes Untuk Mengukur Lulusan Pendidikan Vokasi Yunus Anis; Hersatoto Listiyono; Retnowati; Eko Nur Wahyudi; Sri Mulyani
Servis : Jurnal Pengabdian dan Layanan kepada Masyarakat Vol. 2 No. 2 (2024): Juni
Publisher : CV. Nature Creative Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58641/servis.v2i2.89

Abstract

Pendidikan vokasi bertujuan untuk membekali siswa dengan keterampilan dan pengetahuan yang dibutuhkan untuk siap bekerja di dunia industri. Dalam rangka mengevaluasi efektivitas program pendidikan vokasi, SMK-01 Tonjong Brebes menyelenggarakan pelatihan tracer study bagi para siswanya. Tracer study merupakan metode penelitian yang bertujuan untuk melacak pencapaian lulusan pendidikan vokasi, seperti bekerja, melanjutkan pendidikan, atau berwirausaha. Penelitian ini bertujuan untuk mendeskripsikan program pelatihan tracer study yang dilaksanakan oleh SMK-01 Tonjong Brebes, menganalisis efektivitas program pelatihan dalam membekali siswa dengan keterampilan melakukan tracer study, serta membahas implikasi temuan penelitian untuk peningkatan pendidikan vokasi. Metode penelitian yang digunakan adalah deskriptif kualitatif dengan teknik pengumpulan data melalui observasi, wawancara, dan analisis dokumen. Hasil penelitian menunjukkan bahwa program pelatihan tracer study di SMK-01 Tonjong Brebes cukup efektif dalam membekali siswa dengan keterampilan melakukan tracer study. Temuan penelitian ini menunjukkan bahwa pelatihan tracer study merupakan salah satu strategi yang dapat dilakukan oleh sekolah vokasi untuk meningkatkan efektivitas evaluasi lulusan dan menyesuaikan program pendidikan dengan kebutuhan dunia kerja. Penelitian ini memiliki beberapa implikasi penting bagi kebijakan dan praktik pendidikan vokasi. Penelitian ini memberikan kontribusi yang berharga bagi pengembangan pendidikan vokasi di Indonesia. Dengan menerapkan pelatihan tracer study dan memanfaatkan hasil tracer study secara optimal, sekolah vokasi dapat meningkatkan kualitas lulusannya dan menghasilkan lulusan yang siap kerja dan berdaya saing di dunia industri.
PELATIHAN CETAK SABLON SARING BAGI KELOMPOK USAHA BERSAMA (KUBE) TAMAN PINTAR TUNAS BANGSA, MANYARAN, SEMARANG BARAT Budiarso, Zuly Budiarso; Teguh Khristianto; Hersatoto Listiyono
Servis : Jurnal Pengabdian dan Layanan kepada Masyarakat Vol. 2 No. 2 (2024): Juni
Publisher : CV. Nature Creative Innovation

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

Abstract

Kelompok Usaha Bersama (KUBE) merupakan kelompok pemuda di Taman Pintar Tunas Bangsa Manyaran Semarang Barat. Belum adanya dukungan modal usaha, pemilihan jenis usaha yang sesuai, ketrampilan para usaha di bidang wirausaha yang masih terbatas menjadi alasan tidak berkembangnya kegiatan usaha para remaja KUBE. Untuk mengatasi permasalahan tersebut diperlukan sebuah kegiatan untuk mendukung kegiatan usaha para remaja.Cetak  cetak sablon manual atau sering disebut cetak sablon saring masih menjadi salah satu pilihan dalam dunia percetakan. Pengabdian KepadaMasyarakat (PKM)bertujuanuntuk memberikan ketrampilan di bidang cetak sablon saring bagi remaja di Taman Pintar  Tunas Bangsa. Percetakan sablon merupakan jenis percetakan yang paling sederhana. Proses produksi dan alat produksi yang digunakan sangat sederhana. Dengan kemampuan para remaja di bidang cetak sablon saring dapat dijadikan modal untuk menghidupkan kegiatan usaha .
OPTIMALISASI PENGGUNAAN SCRIBUS UNTUK DESAIN GRAFIS EDITORIAL MAHASISWA TEKNOLOGI REKAYASA MULTIMEDIA GRAFIS Yunus Anis, Yunus Anis; Hersatoto Listiyono; Sri Mulyani; Sunardi; Vici Tiara Anjarsari
Servis : Jurnal Pengabdian dan Layanan kepada Masyarakat Vol. 3 No. 2 (2025): Juni
Publisher : CV. Nature Creative Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58641/servis.v3i2.145

Abstract

Editorial graphic design is a vital skill in the creative industry, especially for producing magazines, brochures, and other publication media. However, the high cost of commercial software licenses such as Adobe InDesign poses a major barrier for students of the Multimedia Graphic Engineering Technology (TRMG) study program in developing editorial design competencies. This community service activity aims to optimize the use of Scribus, an open-source desktop publishing (DTP) software, as a legal and functional alternative for editorial design. The program involved an initial needs assessment, module development, a three-day intensive training, hands-on practice, and pre- and post-training evaluations. The results show a significant improvement in students’ understanding and skills in creating digital publication layouts using Scribus. Additionally, the quality of editorial designs produced by participants improved in terms of visual composition and content structure. This initiative also promotes the ethical and professional use of free-license software. It is expected that Scribus can be integrated into editorial design courses within the TRMG curriculum and support the ongoing development of student portfolios.
Co-Authors Abdul Karim Adelia Aura Diva Zainanda Agus Budi Santosa Agus Perdana Windarto Agus Prasetyo Agus Prasetyo Utomo Agus Prasetyo Utomo Aji Supriyanto Alana Sharfina Wedaningsih Ali Maskur Alif Budi Santoso Anak Agung Istri Sri Wiadnyani Anisa Istiqomah Antono Adhi Askar Yunianto Budiarso, Zuly Budiarso Daniswara, Alfreda Khansa Diah Laila Sani Dwi Agus Diartono Eddy Nurraharjo Eddy Nurraharjo Edi Supriyanto, Edi Edy Supriyanto Edy Winarno Eko Nur Wahyudi Eri Zuliarso Ferdiansyah, Alif Hari Murti Heribertus Yulianton Herny Februariyanti Herny Februariyanti Ibrahim, A. F. Indriani, Vanyariska Isworo Nugroho Isworo Nugroho Jati Sasongko Wibowo Jeffri Alfa Razaq Liana Rahmaziana Muhammad A.R. Hidayat Muji Sukur Novita Mariana P Purwatiningtyas Priambodo, Ilyas Purwadi, Dimas Indra Purwaningtyas , P Purwatiningtyas Purwatiningtyas Purwatiningtyas Purwatiningtyas Purwatiningtyas, P R. Soelistijadi Raharjo, Eddy Nur Ramadhan, Guntur Rani, R.P.S. Retnowati Retnowati Retnowati Retnowati Retnowati Rizal Adi Saputro Rosyida, Elviana S Sunardi S Sunardi Saefurrahman, Saefurrahman Sariyun Naja Anwar Sekar Ayu Ningtyas Sri Mulyani Sri Mulyani Sri Mulyani Sri Yulianto Fajar Pradapa Sugiyamto Sugiyamto Sugiyamto Sugiyamto Sugiyamto Sugiyamto, Sugiyamto Sunardi Sunardi Sunardi sunardi sunardi Sunardi Sunardi Sunardi Sunardi Susi Susilowati, Susi Syamsul Arifin Teguh Khristianto Teguh Khristianto Teguh Kristianto Vici Tiara Anjarsari Vici Tiara Anjarsari Vici Tiara Anjarsari Widiyanto Tri Handoko Yunus Anis Yunus Anis, Yunus Ziana, Liana Rahma Zuly Budiarso Zuly Budiharso