Journal of Applied Computer Science and Technology

Deteksi Helm Keselamatan Menggunakan Jetson Nano dan YOLOv7

Hadi Supriyanto — 2024-02-03

Increasing awareness of the importance of head personal protective equipment in industrial and construction environments has become a major concern in efforts to improve occupational safety. This research developed an early detection system for the use of computer vision-based head protective equipment using the YOLOv7 model and the Jetson Nano controller. The YOLOv7 algorithm was chosen for its ability for fast and accurate object detection. The YOLOv7 model was trained with a total dataset of 2799 images and iterations of 100 epochs to detect head personal protective equipment with a high degree of accuracy. The system captures imagery, activates a warning alarm, and sends a notification to Telegram when a violation occurs on an object that is not wearing a safety helmet. The test results using the confusion matrix method showed that the developed system was able to detect head personal protective equipment with an accuracy rate of 97.23%, which shows the system's ability to recognize personal protective equipment with very high accuracy. In addition, the system also showed a precision value of 98.71% indicating that all detections performed were correct, and a recall of 95.63% which describes the system's ability to recognize most of the head personal protective equipment available. The average FPS result using GPU with CUDA on Jetson Nano reached 5,723 FPS.

Desktop Application for Traceability System on The Printed Circuit Board (PCB) Storage Process

Alvin — 2024-02-03

This paper discusses the development of desktop applications for traceability systems. The application was developed to facilitate data recording and tracking in an electronics manufacturing company's storage process of Printed Circuit Board (PCB) products. The application is developed using the Visual Basic language and Microsoft Excel databases. Additionally, the application is integrated with a barcode scanner to simplify the data entry process from PCBs and employee ID cards. Through the trial process conducted on the developed application, it has generally functioned in accordance with the development goals. Program control validation has been tested through several application access attempts from users registered as operators and administrators. The application has successfully recorded data from inbound and outbound processes, demonstrating storage and tracking functionality. Furthermore, the application has displayed the actual status data of the PCBs present in the warehouse. In terms of user satisfaction, seven users stated that this application was effective and efficient compared to the manual data recording process previously used by the company. This result was obtained from a questionnaire after the application was implemented in the company warehouse.

Analisis Sentimen: Pengaruh Jam Kerja Terhadap Kesehatan Mental Generasi Z

Muhammad Daffa Al Fahreza — 2024-02-03

Mental health is a significant concern in society today, particularly for Generation Z, who are vulnerable to experiencing mental health problems that can disrupt daily productivity. The influence of working hours also contributes to the mental health of this generation. To assess public opinion on this issue, sentiment analysis is needed on social media, especially twitter. This research uses the Gaussian Naïve Bayes algorithm and Support Vector Machine with various stemming algorithms such as Nazief-Adriani, Arifin Setiono, and Sastrawi. The sentiment analysis method is used to assess positive, negative, and neutral sentiment in related tweets. The research results show that the Sastrawi stemming algorithm on the Gaussian Naïve Bayes model achieves 84% precision, 84% recall, and 84% f1-score, with 84% accuracy. Meanwhile, Support Vector Machine achieved 91% precision, 90% recall, 90% f1-score, and 91% accuracy. The Nazief-Adriani stemming algorithm on the Gaussian Naïve Bayes model has 80% precision, 80% recall, and 80% f1-score, with 80% accuracy. Meanwhile, on the Support Vector Machine, precision is 87%, recall is 85%, f1-score is 86%, and accuracy is 85%. Arifin Setiono's stemming algorithm on the Gaussian Naïve Bayes model achieved 81% precision, 81% recall, 81% f1-score, with 82% accuracy, while on Support Vector Machine, 88% precision, 86% recall, 86% f1-score, with 86% accuracy. Public opinion was recorded as 33% positive, 9% neutral, and 58% negative. This research aims to increase public awareness of the importance of mental health, especially regarding the influence of working hours, to create a healthy work environment for Generation Z and society in general, as well as improving the quality of mental health.

Aktivitas Dinamis pada Appreciative Game “Warik the Adventurer” berbasis Finite State Machine

Muhammad Rakha' Naufal — 2024-02-04

Serious games have become potential tools for education due to their advantage of giving a fun experience to the learner. Therefore, game experience is a fundamental element in serious game design. The game experience is mainly produced by the game activity, such as a quest or mission. However, in many serious games, the game activities do not have a clear design and concept, resulting in a poor playing experience which produce poor understanding of the material. Appreciative Game is a game that is based on Appreciative Learning concept. Appreciative Learning concepts could be used to design game activities. Appreciative Learning consists of four main stages. The stages are discovery, dream, design, and destiny. These four stages lay down the foundation of serious game activity. This study uses the Finite State Machine to produce intelligent agents in order to develop more dynamic game activity to enhance the game experience. We developed a 3D game called Warik the Adventurer as the testbed for this research. The game is about the cultural diversity in Indonesia. The game Experience Questionnaire (GEQ) is used to evaluate the player experience. The GEQ resulted in an acceptable score of 3 out of 5.

Chatbot Informasi Penerimaan Mahasiswa Baru Menggunakan Metode FastText dan LSTM

Fahmi Yusron Fiddin — 2024-02-04

New Student Admission (PMB) is an important stage in the continuity of education in an educational institution. The Faculty of Science and Informatics (FSI) at Jenderal Achmad Yani University (UNJANI) provides information services about PMB to prospective students and parents/guardians of prospective students but is still inefficient, so it is necessary to improve PMB information services by using Chatbots as a solution that is able to serve questions effectively and consistent. This study aims to develop a PMB information Chatbot system for FSI using the FastText and Long Short-Term Memory (LSTM) methods. Several methods have been used in Chatbot development research, such as Term Frequency–Inverse Document Frequency (TF-IDF), Bag of Words (BoW), and Convolutional Neural Networks (CNN). However, these studies still have certain limitations, such as the inability to grasp the meaning of words and difficulties in handling certain inputs. In this study, the text classification model uses the FastText method as the stage for representing words in vector form, then combined with several pre-processing methods (Tokenization & Casefolding) and LSTM for the classification stage. Then put it into the Chatbot component according to the architecture that was made. In testing, the Black Box Testing method is used to ensure the functionality of the Chatbot system. The test results show that the Chatbot system is able to understand the topic of questions asked by users properly. The interaction between users and Chatbots also runs smoothly, resulting in appropriate and informative responses. The results of this study are expected to be an effective and consistent solution for providing information about PMB to prospective students and parents/guardians of prospective students at FSI.

Prototipe Deteksi Letak Kebocoran Pipa dengan Optimalisasi Kinerja Penerimaan Paket LoRa menggunakan Pengkodean Parameter Fisik

Dedy Wahyu Herdiyanto — 2024-03-24

The purpose of this research is to determine the effect of the physical coding of LoRa communications on monitoring water pipelines. Optimizing the performance of packet receivers in the LoRa communication system using coding on the physical parameters SF (spreading factor), BW (bandwidth), and CR (coding rate). The detection system consists of 3 sensor nodes, 3 intermediate nodes, and 1 receiver node. Data from these sensors is sent to a cloud database. The SX1278 LoRa communication module works using a 433 MHz frequency. During the transmission process on the LoRa communication system, optimization is carried out for receiving data packets using the parameter coding method of physical spread factors, bandwidth, and coding rate. As a result of the research, it is shown that the greater the value of the third parameter (SF, BW, and CR), such as improvement in packet reception performance, improvement in bit security, and increasing packet resistance to various disturbances in transmission, but the time required for sending data be longer. The optimal parameters for detecting pipe leak locations include SF 10, BW 500 KHz, and CR 4/8. The LoRa SX1278 scenario is optimal with a distance of 400 meters, where packet and byte reception are obtained 100%.

Penerapan Algoritma K-Means Untuk Mengelompokkan Kepadatan Penduduk Di Provinsi DKI Jakarta

Frisma Handayanna — 2024-03-24

DKI Jakarta Province is an attraction for immigrants. If the population increases, if it cannot be resolved and managed well, it will result in bad things such as increasing the number of unemployed and affecting economic growth. Population data is used to help group regions based on population density in DKI Jakarta Province in 2019-2022 using the K-Means clustering method. From the results of the research, it provides a solution for the government to pay attention to population groups with the aim of preventing population density because it causes bad effects, so that community welfare is more guaranteed, so grouping (clustering) of provinces in DKI Jakarta is needed to provide information for people who wish to live in the Province DKI Jakarta. The research proves that the test results carried out clustering iterations of population density data were obtained in three iterations. For the results obtained by calculations using the K-Means method and using the rapidminer application, the results obtained were of the same value, namely the cluster with the highest population density of three districts/cities, namely South Jakarta, East Jakarta and West Jakarta whose population density continues to increase.

Deteksi Clickbait pada Judul Berita Online Berbahasa Indonesia Menggunakan FastText

Muhaza Liebenlito — 2024-03-24

The rise of people accessing news portals has created intense competition between online media to get readers or visitors to maximize their revenue. This is what triggers the development of clickbait. Clickbait can reduce the quality of the news itself, and it also has the potential to be misinformation regarding to news contents as known as fake news. Therefore, it is necessary to detect news titles that contain clickbait. This study aims to obtain an optimal clickbait news title classification model using FastText. To get the optimal model can be done by cleaning the data and optimizing the model's hyperparameters. The model was trained using 9600 training data collected from Indonesian online news. The best model obtained in this study has performance with an accuracy of 77% and an F1-Score of 69%.

Perbandingan Metode Random Forest, Convolutional Neural Network, dan Support Vector Machine Untuk Klasifikasi Jenis Mangga

Ricky Mardianto — 2024-05-12

Mango is a fruit known as the "King of Fruit" due to its rich flavor, vast variability, and high nutritional value. Classifying mangoes based on their external appearance is the initial step in the process of identifying and categorizing mango types conventionally. The classification process can be performed by examining external features such as fruit color, shape, and size. Classifying different types of mango fruits accurately can assist researchers in developing superior varieties and also aid farmers for cultivation purposes, sales, distribution, and selecting the right varieties for local growth and weather conditions. This research conducts the classification of mango types based on color from mango images using machine learning. The study compares three methods, namely Random Forest, Support Vector Machine (SVM), and Convolutional Neural Network (CNN), to determine the best method for classifying mango types based on their images. The dataset underwent preprocessing, where image sizes were standardized to 300 x 300 pixels, and color was changed to grayscale. The dataset was then divided into training and testing data with a ratio of 70:30. Subsequently, the dataset was processed using three methods, and their accuracy results were compared. The findings indicate that the Random Forest method yielded the highest accuracy compared to the other methods, with an accuracy rate of 96%. The accuracy of the SVM method was 95%, and the accuracy of the CNN method was 33%. From these results, it can be concluded that the Random Forest method is highly effective for classifying mango types based on their image compared to SVM and CNN methods.

Pendekatan Metode Ensemble Learning untuk Prakiraan Cuaca menggunakan Soft Voting Classifier

Steven Joses — 2024-06-30

Weather conditions are one of the crucial factors that need attention. Changes in weather conditions significantly impact various activities. Weather condition changes are determined by numerous factors, often occurring within a relatively short period in the atmosphere, such as pressure, wind speed, rainfall, temperature, and other atmospheric phenomena. Issues in weather forecasting arise due to several factors, namely the fluctuating atmospheric conditions. This research proposes the development of a weather forecasting model using the ensemble learning method approach. The weather data used consist of 33746 records with attributes used after preprocessing, namely Temperature, Dew Point, Humidity, Wind Speed, Wind Gust, Pressure, Precipitation, and Condition. Testing in this research employs several single-machine learning methods such as K-Nearest Neighbor (KNN), Logistic Regression, Random Forest, Naive Bayes, and Multi-Layer Perceptron. The Naive Bayes method using default parameters achieves a high accuracy of 99.00%. In the ensemble method, combinations of three methods exhibit excellent accuracy for all combinations. The best combination methods are found in the Soft Voting Classifier method (Random Forest, MLP, Naive Bayes), Soft Voting Classifier (Logistic Regression, MLP, Naive Bayes), and Soft Voting Classifier (Random Forest, KNN, Naive Bayes) with an accuracy of 99.03%.

Pengolahan Citra Berbasis Video Proccesing dengan Metode Frame Difference untuk Deteksi Gerak

Yovi Apridiansyah — 2024-06-30

This study discusses the detection of motion of objects in video by utilizing the Frame difference method which aims to process video so as to produce Frames on moving objects. The use of mobile cameras produces video data that is used as test data, the test data is processed with the Frame difference method so as to produce a number of Frames on moving objects in order to detect moving objects in the video because the function of this method is a form of video background reduction that is simplified by a number of pixels in the video. This method process is based on the difference between two consecutive frames in the video aimed at finding differences that occur during the detection process. When processed for detection, the absolute value in the pixel is greater than the predetermined threshold value, it will be considered as a moving object, so that the detection results from the motion detection process will form a box object on the moving object. In this study, the test data used used 20 video data samples with descriptions, 10 test data with bright quality (daytime) and 10 unlit test data (night) with the aim of being able to see how much the level of performance accuracy of the Frame difference method. The test results obtained 16 out of 20 test data that were successfully detected correctly (True Positive), there were 2 test data that resulted in a False Positive error, and 2 test data that resulted in a False Negative error. This shows that the Frame difference method can provide a fairly high level of accuracy in detecting moving objects in the video. The percentage level of accuracy with confussion matrix testing has a precission value of 88%, recal 88% and an accuracy value of 80%.

Identifikasi Tujuan Tata Kelola Teknologi Informasi PLT FST UIN Jakarta Menggunakan Framework COBIT 2019

Nur Aeni Hidayah — 2024-06-30

Information technology (IT) has an important role in increasing efficiency and providing benefits to decision making in various sectors. IT governance is important for organizational sustainability and performance. This research focuses on the Integrated Laboratory Center (PLT) FST UIN Jakarta which still depends on external parties for IT governance, creating potential risks and uncertainty in its operations. Therefore, this research aims to analyze the implementation of COBIT 2019 as a guide in identifying IT governance objectives that suit the needs of PLT FST UIN Jakarta. The research method used in this research involves the COBIT 2019 theoretical approach, with a focus on the goal cascade concept. This approach enables a systematic translation process of stakeholders' needs for IT systems into more specific goals and appropriate strategic steps. The results show eight COBIT 2019 domains that are relevant to IT governance objectives in PLT, which include APO02 (Managed strategy), APO08 (Managed relationships), APO10 (Managed vendors), APO11 (Managed quality), BAI04 (Managed availability and capacity), DSS01 (Managed operations), DSS03 (Managed problems(, and MEA02 (Managed system of internal control) The recommendations resulting from this research are designed to improve the performance and maturity of IT processes at PLT FST UIN Jakarta in accordance with the IT governance objectives that have been identified. This is expected to provide a stronger foundation for managing and optimizing the use of IT to support operations and achieve organizational goals more effectively.

Implementasi Re-design Application Mobile MRT Jakarta Menggunakan Metode User Centered Design

Diana Rahma Fahriyah — 2024-06-30

MRT Mobile Application is used to access information easily and quickly for MRT users, such as schedule information, routes, stations, fares, and other additional services. Based on observations and questionnaires regarding the MRT mobile application, users often experience difficulties or obstacles in interacting with the application interface. Starting with a complex interface, confusing navigation of features to difficulty finding the information you need quickly. This affects the use of the application and reduces the level of user satisfaction. Based on these conditions, it is necessary to redesign the user interface of this application to make it easier to use. This research uses User Centered Design (UCD) method with four phases, which is Understand Context of Use, Specify User Requirements, Design Solution, and Evaluation Against Requirements in analyzing the UI/UX design process of the MRT mobile application. In the design stage, UCD principles are used to integrate insights from user data into UI/UX design process based on results of questionnaire analysis. The final result of this research is a prototype design and user interface design for the MRT mobile application that meets the expectations of application users.

Klasifikasi Metode Data Mining untuk Prediksi Kelulusan Tepat Waktu Mahasiswa dengan Algoritma Naïve Bayes, Random Forest, Support Vector Machine (SVM) dan Artificial Neural Nerwork (ANN)

Satrio Junaidi — 2024-06-30

Timely graduation of students is essential for determining the quality of college. Universities must know the percentage of students' ability to complete their studies on time. So, to deal with this problem, data mining classification is carried out to predict student graduation on time to find patterns for student on-time graduation predictions. This research can yield new information to help colleges anticipate student graduations that are not on time. The method used is a classification data mining method with 4 algorithms: naïve Bayes, random forest, support vector machine (SVM), and artificial neural network (ANN). The attributes used are gender, parental income, length of guidance, working student status or not, semester 1 to semester 8 grades, and GPA. This study used Python 3 programming language on jupyter notebooks in Anaconda to process datasets. The distribution of datasets is divided by 70% for training data and 30% for testing data. The results of this study were obtained with the best algorithm accuracy in the support vector machine (SVM) algorithm is 0.94. Based on the results of this study, the accuracy is good for predicting student graduation on time.

Analisis Algoritma Klasifikasi Untuk Mengidentifikasi Potensi Risiko Kesehatan Ibu Hamil

Jajang Jaya Purnama — 2024-06-30

The health of pregnant women has an important aspect in efforts to achieve the birth of a healthy baby. So early detection of the health of pregnant women has important. In this study the author identified potential maternal health risks for pregnant women by classifying them used machine learning which aims to analyze maternal health datasets with several algorithms including Random Forest, Extra Trees, Extreme Gradient Boosting, Decision Tree, and Light Gradient Boosting Machine. From several classification results carried out analysis and evaluation shown that the Random Forest classification algorithm provided optimal performance with an accuracy of 82,15%. These findings confirmed that the model created could identify complex patterns and relationships between features relevant to the classification of potential health risks for pregnant women at high, medium and low levels. These results have important implications in maternal care, because they cann help doctors and medical personnel make more appropriate and effective decisions in dealing with maternal health risks and provide insight into pregnant women from an early age regarding their health conditions.