Garuda - Garba Rujukan Digital

Three layer hybrid learning to improve intrusion detection system performance Harwahyu, Ruki; Erasmus Ndolu, Fajar Henri; Overbeek, Marlinda Vasty
International Journal of Electrical and Computer Engineering (IJECE) Vol 14, No 2: April 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijece.v14i2.pp1691-1699

In imbalanced network traffic, malicious cyberattacks can be hidden in a large amount of normal traffic, making it difficult for intrusion detection systems (IDS) to detect them. Therefore, anomaly-based IDS with machine learning is the solution. However, a single machine learning cannot accurately detect all types of attacks. Therefore, a hybrid model that combines long short-term memory (LSTM) and random forest (RF) in three layers is proposed. Building the hybrid model starts with Nearmiss-2 class balancing, which reduces normal samples without increasing minority samples. Then, feature selection is performed using chi-square and RF. Next, hyperparameter tuning is performed to obtain the optimal model. In the first and second layers, LSTM and RF are used for binary classification to detect normal data and attack data. While the third layer model uses RF for multiclass classification. The hybrid model verified using the CSE-CIC-IDS2018 dataset, showed better performance compared to the single algorithm. For multiclass classification, the hybrid model achieved 99.76% accuracy, 99.76% precision, 99.76% recall, and 99.75% F1-score.

U-TAPIS Sal-Tik : Typing Error Detection Using Random Forest Algorithm Overbeek, Marlinda Vasty; Glennardy, Bryan; Mediyawati, Niknik; Nusantara, Samiaji Bintang; Sutomo, Rudi
ULTIMATICS Vol 16 No 1 (2024): Ultimatics : Jurnal Teknik Informatika
Publisher : Faculty of Engineering and Informatics, Universitas Multimedia Nusantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31937/ti.v16i1.3563

The development of technology in the field of journalism has grown very rapidly. However, in the field of journalism there are still frequent deviations from the language on online news portals. This can be seen from the aspect of spelling and word usage. Spelling mistakes that occur in the news can cause the information contained in the news to be unclear and ambiguous. Based on these problems, a study was conducted to create a model to detect type error in Indonesian. This model is created using the random forest algorithm. random forest is an algorithm that works by building several decision trees and then combining the decisions from each tree that has been built and taking the most votes from the predictions of each tree so that it will produce stable and accurate predictions. The results of the accuracy of the model in the research that has been done is 100%. However, it should be noted that this 100% result is that the model is able to detect words that are already contained in the dataset. Based on the evaluation results that have been carried out, because the detected word is contained in the dataset, the accuracy issued is 100%. The built model successfully detects type error in Tribunnews news articles.

Sistem Identifikasi Titik Kritis Halal Menggunakan Algoritma Forward Chaining Hin, Alexander Moya; Kusnadi, Adhi; Overbeek, Marlinda Vasty; Prawira, Oqke; Khaeruzzaman, Yaman; Prasetya, Syarief Gerald
JURNAL Al-AZHAR INDONESIA SERI SAINS DAN TEKNOLOGI Vol 8, No 1 (2023): Januari 2023
Publisher : Universitas Al Azhar Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36722/sst.v8i1.1285

Halal products are obligatory to be used by people who are Muslim. When viewed in terms of the number of the Muslim population in the world and Indonesia, halal products have very potential economic opportunities. However, halal products have the risk of becoming non-halal if the accompanying process and storage do not follow halal rules. Therefore, it is necessary to identify the critical halal point, the point where the potential for such change occurs. So far, identification is made manually, of course there will be opportunities for identification errors to happen and it will take a relatively long time. To overcome these problems, identification can use a computer-based system. Forward chaining is an algorithm that is suitable for identifying halal points, because in SJH LPPOM MUI there is a decision tree for identifying halal critical points which is carried out in the same forward sequence as the forward chaining algorithm process flow. The development of a halal critical point identification system is carried out using the Software Development Life Cycle V-model method, the PHP programming language and the MySQL Database Management System. The system was successfully tested using Whitebox testing, including unit testing, integration testing, and overall system testing. Then testing using Blackbox testing techniques by comparing the results of identifying critical points using the system with the results of identifying critical points manually producing the same results. User satisfaction testing was also carried out using the End User Computing Satisfaction method and obtained an average satisfaction score of 86.53%Keywords – halal products, critical halal point, AI, forward chaining

WEB-BASED DESIGN OF DHARMA WANITA ASSOCIATION (DWP) LPKA CLASS II JAKARTA FINANCIAL REPORT APPLICATION Tobing, Fenina Adline Twince; Surbakti, Eunike Endariahna; Overbeek, Marlinda Vasty
Jurnal Sinergitas PKM & CSR Vol. 6 No. 2 (2022): OCTOBER
Publisher : Universitas Pelita Harapan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19166/jspc.v6i2.6360

Lembaga Pembinaan Khusus Anak (LPKA)Class II Jakarta, there is a Dharma Wanita Association (DWP) Jakarta Class II LPKA organization whose members are all the wives of civil servants and also female civil servants who serve in that place. One of the keys to the success of an organization is being able to manage finances well, one of which is reflected in routinely making financial reports every month. Not only important for large companies, financial reports that are compiled and updated regularly are also important for MSME businesses, online businesses or organizations. It doesn't matter if someone doesn't have an educational background in accounting, but is still able to make complete financial reports.Technological advances encourage DWP LPKA Class II Jakarta to be able to carry out their work in terms of financial reporting to make it easier and more efficient by using applications that we have researched and implemented through the Multimedia Nusantara University (UMN) Community Service Program (PKM) in 2022 and DWP LPKA Class II Jakarta members received this form of community service with great enthusiasm. Through this PKM activity it is hoped that there will be an increase in the application of science and technology in society (mechanisms, IT, management) and for improving community values from a social and financial management perspectiveabstract in bahasaLembaga Pembinaan Khusus Anak (LPKA) Kelas II Jakarta, terdapat organisasi Dharma Wanita Persatuan (DWP) LPKA Kelas II Jakarta yang beranggotakan seluruh Istri PNS dan juga PNS wanita yang bertugas di tempat tersebut. Salah satu kunci sukses sebuah organisasi adalah mampu mengelola keuangan dengan baik, salah satunya tercermin dari rutin membuat laporan keuangan setiap bulan. Tidak hanya penting bagi perusahaan besar, laporan keuangan yang disusun dan diperbaharui secara berkala juga penting dilakukan oleh bisnis UMKM, usaha online atau organisasi. Tidak masalah jika seseorang tidak memiliki latar belakang pendidikan akuntansi, namun tetap mampu membuat laporan keuangan yang lengkap.Kemajuan Teknologi mendorong untuk membantu DWP LPKA Kelas II Jakarta untuk dapat melaksanaan pekerjaan mereka dalam hal pelaporan keuangan agar lebih mudah dan efisien dengan menggunakan aplikasi yang telah kami teliti dan kami implementasikan melalui Program Pengabdian Kepada Masyarakat (PKM) Universitas Multimedia Nusantara (UMN) tahun 2022 dan anggota DWP LPKA Kelas II Jakarta memerima bentuk pengabdian masyarakat yang kami lakukan dengan sangat antusias. Melalui kegiatan PKM ini diharapkan adanya peningkatan penerapan IPTEK di masyarakat (mekanisme, IT, manajemen) dan untuk perbaikan tata nilai masyarakat dari segi sosial dan manajeme keuangan.

Ensemble Learning - Random Forest Algorithm to Classify Obesity Level Abigail, Tesalonika; Overbeek, Marlinda Vasty
IJNMT (International Journal of New Media Technology) Vol 11 No 2 (2024): Vol 11 No 2 (2024): IJNMT (International Journal of New Media Technology)
Publisher : Universitas Multimedia Nusantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31937/ijnmt.v11i2.3709

Obesity is one of the serious global health problems caused by excessive accumulation of body fat. According to the World Health Organization (WHO), the prevalence of obesity has tripled in the last 40 years, with 650 million out of 1.9 billion overweight adults suffering from obesity. Obesity is a non-communicable disease that increases the risks of more dangerous diseases, such as heart disease and cancer. Therefore, early detection of obesity level is crucial. Currently, Body Mass Index (BMI) serves as a measurement indicator, but it tends to overestimate obesity for those with high muscle mass and vice versa, making it ineffective as it only relies on height and weight, without considering body composition and daily activities. To solve this, the best Random Forest model has been developed, selected based on the results of model selection after comparisons using feature selection and hyperparameter tuning. The selected model successfully improved accuracy by 1.4%, which then implemented into a web-based system to classify obesity levels. Evaluation of the model resulted in Precision, Recall, F1-Score, and Accuracy of 97%, 97%, 97%, and 96.8% respectively. Based on these evaluation results, it can be concluded that this system is highly effective in classifying obesity levels.

Using Convolutional Neural Network and Saliency Maps for Cirebon Batik Recognition Aditiya, Yoga; Overbeek, Marlinda Vasty; Pomalingo, Suwito
ULTIMATICS Vol 17 No 1 (2025): Ultimatics : Jurnal Teknik Informatika
Publisher : Faculty of Engineering and Informatics, Universitas Multimedia Nusantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31937/ti.v17i1.4026

Cirebon Batik is one of Indonesia's cultural heritages that has its own unique patterns and motifs, reflecting the cultural richness and history of its region of origin. This study aims to address the challenges in classifying the complex motifs of Cirebon Batik by implementing Convolutional Neural Network (CNN) and Saliency Map methods. The three main motifs used are Mega Mendung, Singa Barong, and Keratonan. The dataset was obtained from various online sources and processed using image augmentation techniques. CNN is used to recognize complex visual patterns, while Saliency Map highlights important areas in the image that influence the model's decision. The results show that the developed CNN model achieved an accuracy of 82%, precision of 83%, recall of 82%, and F1-score of 82%. The use of Saliency Map provides better interpretability and enhances the understanding of the classification process

Multichannel Slotted ALOHA Simulator Design for Massive Machine-Type Communication (mMTC) on 5G Network Feliana, Ferlinda; Harwahyu, Ruki; Overbeek, Marlinda Vasty
International Journal of Electrical, Computer, and Biomedical Engineering Vol. 1 No. 2 (2023)
Publisher : Universitas Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62146/ijecbe.v1i2.8

Massive Machine-type Communication (mMTC) is one of the main service scenarios in 5G. At the time of initializing the connection to the base station, the MTC machines will make a connection request via the random access procedure. One of the schemes of random access procedure for handling this connection request is similar to how multichannel slotted ALOHA works. Multichannel slotted ALOHA itself is a development of the slotted ALOHA scheme which originally has only a single channel. At the initial state of mMTC, there will be an explosion of the number of demands to the available channels. Given the number of machines that will be connected, the likelihood of a collision on the same channel increases. As a result, the probability of failure also increases. The system's configuration has an impact on the likelihood of success and the time it takes to achieve it. The number of channels influences the likelihood of collisions, the backoff window influences the transmission distribution in each slot, and the maximum transmission limits the ability of device retransmission. These three arrangements have an impact on one another. The simulator build in this research is expected to make it easier for researchers to optimize multichannel slotted ALOHA configurations in 5G to handle the surge in access demands from mMTC devices.

SISTEM REKOMENDASI DESTINASI WISATA DI KOTA KUPANG DENGAN METODE WEIGHTED PRODUCT Overbeek, Marlinda Vasty; Naatonis, Remerta Noni
HOAQ (High Education of Organization Archive Quality) : Jurnal Teknologi Informasi Vol. 10 No. 1 (2019): Jurnal HOAQ - Teknologi Informasi
Publisher : STIKOM Uyelindo Kupang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52972/hoaq.vol10no1.p30-34

From 2006 to the latest data collection in 2017, the number of domestic and foreign tourists visiting East Nusa Tenggara is increasing. In Kupang city as the provincial capital of East Nusa Tenggara. Kupang City is not just a temporary stop when tourists want to transit to other islands in East Nusa Tenggara but have become tourist destinations. The increasing number of tourist visits is not accompanied by a system of recommendations about tourist destinations in the city of Kupang, In this study a recommendation system was built in Kupang City using the weighted product method, which is one of the techniques of Multiple Alternative Decision Making. Weighted products use multiplication to connect attribute ratings. The criteria used in this study are 3 criteria, namely costs, facilities provided and reviews from previous visitors. While for locations or tourist destinations there are 55 tourist destinations in Kupang City which are the alternatives in this study. From the results of the study, for natural attractions, Namosain Beach ranked first to be visited with a value of 0.00519. As for artificial tourism objects, Nostalgia Park is the first place with a value of 0.0805. As for culinary tourism, Nostalgia Park culinary ranks first at 0.0904.

HISTOGRAM OF ORIENTED GRADIENT UNTUK DETEKSI EKSPRESI WAJAH MANUSIA Overbeek, Marlinda Vasty
HOAQ (High Education of Organization Archive Quality) : Jurnal Teknologi Informasi Vol. 10 No. 2 (2019): Jurnal HOAQ - Teknologi Informasi
Publisher : STIKOM Uyelindo Kupang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52972/hoaq.vol10no2.p81-86

This research focuses on the detection of human facial expressions using the Histogram of Oriented Gradient algorithm. Whereas for the classification algorithm, Convolutional Neural Network is used. Image data used in the form of seven different expressions of humans with the extraction of 48x48 pixels. The use of Histogram of Oriented Gradient as a feature extracting algorithm, because Histogram of Oriented Gradient is good to be used in detecting moving objects. Whereas Convolutional Neural Network is used because it is an improvement of the Multi Layer Perceptron algorithm. Of the three epoches done, it produced the best accuracy of 77% re-introduction of human facial expressions. These results are quite convincing because it only uses three epochs.

Ensemble Learning - Random Forest Algorithm to Classify Obesity Level Abigail, Tesalonika; Overbeek, Marlinda Vasty
IJNMT (International Journal of New Media Technology) Vol 11 No 2 (2024): Vol 11 No 2 (2024): IJNMT (International Journal of New Media Technology)
Publisher : Universitas Multimedia Nusantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31937/ijnmt.v11i2.3709

Obesity is one of the serious global health problems caused by excessive accumulation of body fat. According to the World Health Organization (WHO), the prevalence of obesity has tripled in the last 40 years, with 650 million out of 1.9 billion overweight adults suffering from obesity. Obesity is a non-communicable disease that increases the risks of more dangerous diseases, such as heart disease and cancer. Therefore, early detection of obesity level is crucial. Currently, Body Mass Index (BMI) serves as a measurement indicator, but it tends to overestimate obesity for those with high muscle mass and vice versa, making it ineffective as it only relies on height and weight, without considering body composition and daily activities. To solve this, the best Random Forest model has been developed, selected based on the results of model selection after comparisons using feature selection and hyperparameter tuning. The selected model successfully improved accuracy by 1.4%, which then implemented into a web-based system to classify obesity levels. Evaluation of the model resulted in Precision, Recall, F1-Score, and Accuracy of 97%, 97%, 97%, and 96.8% respectively. Based on these evaluation results, it can be concluded that this system is highly effective in classifying obesity levels.

Title

Found 11 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 11 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 11 Documents
Search