Claim Missing Document
Check
Articles

Found 5 Documents
Search

APPLICATION AND ATTRIBUTE ANALYSIS IN THE MODEL OF CLASSIFYING HEART DISEASE Rosdiana Rosdiana; Vera Novalia; Hafizh Al Kautsar Aidilof; Muhammad Danil; Muhammad Ikhsanul Fikri
Multica Science and Technology Vol 1 No 2 (2021): Multica Science and Technology
Publisher : Universitas Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47002/mst.v1i2.280

Abstract

The heart is the central center in the human circulatory system. A malfunction of the heart that is not functioning is a condition in which the heart cannot carry out its duties properly. Selection of features that can reduce a very large dataset and in a data set that is not suitable can use a reduction model. The classification process is strongly influenced by an attribute. Various types of inappropriate redundancy have a negative effect on classification accuracy. Heart disease data was taken from the UCI Machine Learning Repository dataset. In this study, the researchers used the K-Nearest Neighbor (KNN) algorithm where the K-Nearest Neighbor algorithm can classify the results of heart disease accurately. The results are as follows 1.67358 rank one 1.33949 rank two, 1.27260 rank three, 1.2528 rank four, 1.24193 rank last
Design and Implementation of an RFID-Based Automatic Doorstop System with Website and Telegram Bot Integration Zainul Anwar Adi Putra; Rizal Tjut Adek; Hafizh Al Kautsar Aidilof
Tech-E Vol. 8 No. 2 (2025): TECH-E (Technology Electronic)
Publisher : Fakultas Sains dan Teknologi-Universitas Buddhi Dharma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31253/te.v8i2.3447

Abstract

This research develops a prototype of an automatic doorstop control system based on Radio Frequency Identification (RFID) and the Internet of Things (IoT) integrated with a website-based information system and Telegram bot. This system is specifically designed to improve efficiency and security in access management at Malikussaleh University, by overcoming the vulnerabilities and limitations of traditional manual access control systems that are prone to security risks. The system uses RFID sensors to read user identity cards as access verification, while infrared (IR) sensors detect objects near the door to ensure security during automatic door operation. The system has an easy-to-use web interface for efficient management of data and activity records. In addition, real-time notifications are sent via Telegram bot to provide administrators with detailed information on access attempts. Tests show that the RFID sensor is capable of accurately reading ID cards at distances of up to 2 cm, while the IR sensor detects objects near the door quickly and precisely. The servo motors used had an average response time of 2 seconds to open and close the door. With a 98% accuracy rate on the RFID sensor, this system provides a reliable solution for automatic access control needs. With the advantages of high accuracy, fast response, and ease of integration, this prototype is expected to be implemented in various educational institutions and other public facilities.
Performance Evaluation of ARIMA Model in Forecasting Rice Production Across Sumatera, Indonesia Imam Rosadi; Muhammad Fikry; Hafizh Al kautsar Aidilof
Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN) Vol. 2 (2024): Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN)
Publisher : Faculty of Engineering, Malikussaleh University

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

Abstract

Abstract In this paper, we present a comprehensive performance evaluation of the ARIMA (AutoRegressive Integrated Moving Average) model in forecasting rice production across Sumatera, Indonesia. Rice is a crucial staple crop, feeding more than half of the global population. In Sumatera, rice plays a vital role in food security, yet its cultivation is highly dependent on specific environmental conditions such as temperature, humidity, and rainfall. This study leverages historical time-series data from the years 2000 to 2020, collected from eight key provinces: Aceh, North Sumatera, West Sumatera, South Sumatera, Riau, Jambi, Bengkulu, and Lampung. The objective is to forecast rice production for the years 2021-2024 using the ARIMA method. Through rigorous model selection and evaluation, ARIMA (3,0,2) was identified as the most suitable model, providing accurate forecasts with a Mean Squared Error (MSE) of 0.0325 and a Mean Absolute Error (MAE) of 0.1445. These low error rates demonstrate the model’s capacity to capture the inherent fluctuations in rice production trends across Sumatera. The findings offer critical insights for future rice production trends and can guide policy-makers in formulating effective food security strategies. This research contributes significantly to the understanding of rice production dynamics and the application of ARIMA models in agricultural forecasting. Keywords: Rice production; Mean Squared Error; Mean Absolute Error; ARIMA; Sumatera.
Smart Fire Prevention: An IoT Approaceh To Detecting LPG Leaks And Fire Hazards Luthvy Ilhamdi; Muhammad Fikry; Hafizh Al Kautsar Aidilof
Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN) Vol. 2 (2024): Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN)
Publisher : Faculty of Engineering, Malikussaleh University

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

Abstract

In this paper, we address the serious fire risks posed by Liquefied Petroleum Gas (LPG) leaks, which can lead to significant material damage and loss of life. These incidents are often caused by human error or the absence of effective early warning systems capable of timely leak detection. To tackle this issue, we have developed an automatic gas leak detection system integrated with the Internet of Things (IoT). The system utilizes the ESP32 microcontroller as the main control unit, along with an MQ2 gas sensor for detecting LPG leaks and a fire sensor for identifying fire hazards. Additional components include a fan to enhance air circulation in case of gas accumulation and an automatic water pump that activates upon fire detection, aiding in prompt fire extinguishment. The system is also equipped with an LCD to display real-time gas levels in the environment, providing visual feedback to users. For enhanced functionality, this system connects to the Blynk application, allowing remote monitoring and control via smartphone. This feature enables users to receive instant notifications upon detecting gas leaks or fires and to manually control the fan or water pump if necessary. The primary objective of this system is to provide early detection and automatic response to gas leaks and fire hazards, thereby reducing the risk of fire-related accidents. This IoT-based approach offers a reliable solution to enhance safety by ensuring rapid responses to gas leaks and fires, ultimately minimizing damage and protecting lives.
Face Recognition System For Student Identification Using VGG16 Convolutional Neural Network Chrisnata Manihuruk; Muhammad Fikry; Hafizh Al Kautsar Aidilof
Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN) Vol. 2 (2024): Proceedings of International Conference on Multidisciplinary Engineering (ICOMDEN)
Publisher : Faculty of Engineering, Malikussaleh University

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

Abstract

In this paper, we present a robust facial recognition system designed to identify students at Department of Informatics in Universitas Malikussaleh using a Convolutional Neural Network (CNN) algorithm, specifically the VGG16 architecture. The advancement of information technology and machine learning has significantly improved facial recognition capabilities, establishing it as a reliable alternative to traditional identification methods such as fingerprinting and iris scanning. Our approach leverages a diverse dataset captured from five different angles, enhancing the representation of facial features and improving model training. The system development comprises several critical stages, including image acquisition, preprocessing, model training with training and validation data, and performance evaluation. Experimental results indicate that the CNN model achieves an impressive accuracy of 99.09% on training data and 100% on both validation and testing datasets. These findings affirm the model's high classification accuracy across the tested classes, underscoring the effectiveness of the VGG16-based CNN in facial recognition applications. The implications of this study suggest that the developed system can significantly enhance digital attendance and security systems, catering to the growing demand for reliable AI-driven security technologies in contemporary society. We anticipate that with its promising outcomes, this system can be implemented on a larger scale, contributing to the ongoing advancement of AI-based security solutions.