JITCE (Journal of Information Technology and Computer Engineering) 2022-01-16T00:16:17-05:00 Editor JITCE Open Journal Systems <p><strong>JITCE (Journal of Information Technology and Computer Engineering)</strong>&nbsp;is a scholarly periodical. JITCE will publish research papers, technical papers, conceptual papers, and case study reports. This journal is published by<a href=""> Computer System Department</a> at&nbsp;<a href="" target="_blank" rel="noopener">Universitas Andalas</a>, Padang, West Sumatra, Indonesia.</p> <p>One volume of JITCE consisted of two editions, which are published in March and September each year. Articles are written in Bahasa Indonesia (Indonesian language) OR English. Abstracts&nbsp;<strong>must be in English</strong>.</p> Sistem Pendeteksi Gejala Awal Tantrum Pada Anak Autisme Melalui Ekspresi Wajah Dengan Convolutional Neural Network 2022-01-15T23:55:16-05:00 Nefy Puteri Novani Dini Ramadhani Salsabila Ratna Aisuwarya Lathifah Arief Nelia Afriyeni <p>Tantrums are outbursts of anger and they can occur at any age. An attitude tantrum or what is commonly referred to as a temper tantrum is a child's outburst of anger that often occurs when a child shows negative behavior. Emotional outbursts of tantrums that occur in children with autism are not only to seek the attention of adults, but also as an outlet for a child's feelings for parents and those around him on a whim or feeling he is feeling, but the child cannot convey it. For this reason, researchers propose a system for detecting early symptoms of tantrums in children with autism through facial expressions with CNN. The CNN method is one of the deep learning methods that can be used to recognize and classify an object in a digital image. Then the preprocessing process is carried out using labeling on the data. Then the CNN architecture is designed with input containing 48x48x1 neurons. The data was then trained using 357 epochs with an accuracy rate of 72.67%%. Then tested using test data for children with autism to get an average accuracy value of 72.67%%.</p> 2022-01-15T23:50:02-05:00 ##submission.copyrightStatement## Sistem Kendali Sirkulasi Udara dan Pembatasan Jumlah Pelanggan Toko Berbasis IoT 2022-01-13T22:45:57-05:00 Labiq Al Hanif Aditya Putra Perdana Prasetyo Huda Ubaya <p>The emergence of the COVID-19 pandemic in early 2020 had a major impact on human life on a global scale. Many actions and policies are aimed at anticipating transmission and breaking the chain of the spread of the COVID-19 virus, thus requiring store owners to implement various health protocols. This study discusses the monitoring system for the condition of the storeroom in real-time with the IoT concept, and the implementation of Sugeno fuzzy logic in controlling the speed of the exhaust fan motor to circulate air in the room and limit the number of customers during the COVID-19 pandemic based on conditions of temperature, humidity, and many people in the storeroom. The actual test results from the implementation of Sugeno <em>fuzzy</em> logic show that the system has good performance in controlling the speed of the exhaust fan and limiting the number of customers based on the level of danger of the potential COVID-19 transmission in the room automatically and can monitor the condition of the room through the website in real time.</p> 2021-12-31T08:10:29-05:00 ##submission.copyrightStatement## Implementasi Cloud Based Video Conference System Menggunakan Amazon Web Service 2021-12-31T16:36:51-05:00 Alde Alanda Deni Satria <p>Since December 2019, the world and Indonesia have fought a major disaster, namely the Covid-19 virus pandemic. With the rapid spread or transmission of the virus, the Indonesian government decided to impose social distancing or social restrictions that impacted the education sector. Students and lecturers cannot conduct lectures face-to-face in class or laboratory, but lectures must be conducted online. For that, we need an open-source system developed by the campus in carrying out online courses. This application was developed using cloud technology and JITSI as an open-source video-conferencing application. In this study, testing of the features that exist in video conferencing and resource usage on the server is carried out. The results of feature testing on the application run as expected with several important features used for learning such as chat, share screen, recording features that can run optimally. The result tested the system resources based on the number of participants, 31 participants with an average use of 2.1GB RAM and 78 participants with an average RAM usage of 2.8GB.</p> 2021-12-31T07:15:15-05:00 ##submission.copyrightStatement## Rancang Bangun Alat Pendeteksi NOx dan CO Berbasis Mikrokontroler ESP32 dengan Notifikasi Via Telegram dan Suara 2021-10-01T08:16:03-04:00 Mutiara Asmazori <p>The design of NOx and CO detectors based on notifications via telegram and voice has been carried out. This detector consists of a gas sensor MQ-135 as a nitrogen oxide detector and an MQ-7 sensor as a carbon monoxide detector. Data processing is carried out using an ESP32 microcontroller which can send results to a telegram bot and play sound using speakers connected to the ISD 1820 sound module. The tool made can send notifications if the concentration of nitrogen oxides and carbon monoxide exceeds 50 ppm. The test is carried out by burning waste to produce smoke. Burning smoke contains various gases and particles that are harmful to the body. The characterization of the MQ-135 sensor was carried out by comparing the data obtained from the ISPU to measure nitrogen oxide gas and producing an error value of 9.09%. Meanwhile, the characterization of the MQ-7 sensor was carried out using a biogas analyzer and resulted in an error ratio of 3.26%. These results prove that the tools that have been designed can work well.</p> 2021-09-30T00:00:00-04:00 ##submission.copyrightStatement## Metode Kernel Distance Classifier Terhadap Klasifikasi Penyakit Jantung 2022-01-16T00:16:17-05:00 Kasiful Aprianto <p>This study compares the Support Vector Machine (SVM) and Kernel Distance Classification (KDC) methods to classify heart disease. SVM works by transforming data into higher dimensions using the kernel and classifying data linearly using a hyperplane. Meanwhile, KDC works by finding points that represent each classification from the data that has been transformed into a higher dimension using the kernel, and the new data is predicted based on the closest distance from the point of each classification. The results show that the accuracy produced by SVM is 81.11%. The accuracy produced by the SVM model is better than that produced by the KDC model of 80.47% with a difference of 0.64%, even though both models use kernel transformation.</p> 2021-09-30T00:00:00-04:00 ##submission.copyrightStatement##