Jurnal Informatika: Jurnal Pengembangan IT
Not a member yet
437 research outputs found
Sort by
Analisis Sentimen Pembangunan IKN pada Media Sosial X Menggunakan Algoritma SVM, Logistic Regression dan Naïve Bayes
No full textsocial media X or formerly more familiar with Twitter is one of the familiar social media and has many users in the world whis is a platform for accesing some information and commenting both suggestions and criticsm related to the development of the Capital City of the Archipelago (IKN) which is the center of smart government in East Kalimantan. There are indormation, suggestions and criticisms addressed to the @ikn_id account directly addressed to the Indonesia government as well as public opinions related to IKN by using the IKN hashtag. Public sentiment on the issue is in the form of text on IKN Development. This research aims to analyze public opinion on the government's decision to build the Capital City of Nusatara (IKN) conveyed through X social media using appropriate data analysis methods by comparing the performance of support vector machine, logistic regression, and naïve bayes algorithms and identifying the most effective algorithm in sentiment analysis. The method used in this research to analyze sentiment are support vector machine, logistic regression and naïve bayes. The use of these three algorithms is also to compare the accuracy that is better than other algorithms. The results obtained using the Support Vector Machine algorithm is 80% while using the Logistic Regression and naïve bayes algorithms are 79%
Implementation of Radio Frequency Identification in Student Presence Applications with Multi Social Media Notification
No full textRadio-Frequency Identification or commonly called RFID is a technology that can be integrated into various softwares to increase operational efficiency and effectiveness. In an educational unit, one of the services that needs to be repaired or improved is recording student attendance. However, recording and monitoring student attendance in almost all educational units in Indonesia, and especially in Tegal City is still carried out conventionally, so that educational units cannot provide feedback to parents about their children's attendance in real time. This condition is the main basis for conducting this research. In this research, RFID technology was implemented into an desktop based application with the aim of making it easier for educational units to record student attendance automatically, and assisting schools in providing feedback about student attendance to parents through social media services (whatsapp or telegram) and increasing enthusiasm students in taking attendance. The method chosen for development is the waterfall method, this method ensures that all stages are carried out sequentially. The research application has been tested using the black box testing method, the test results indicate that the application functionality is running well
Model Perilaku Pasien Pada Aplikasi Berbasis Kesehatan Menggunakan Metode Design Thinking
No full textThe COVID-19 pandemic has had a significant impact on health, economy, and society in Indonesia. In dealing with the pandemic, the term New Normal emerged, namely a change in behavior to continue normal activities by implementing health protocols. One of the innovations that has developed in this digital era is a health-based application designed to increase the accessibility of health services, such as online consultations, health monitoring, medication reminders, and health education. However, the success of this application depends not only on technology, but also on a deep understanding of patient behavior as users. Understanding patient needs, preferences, and challenges is important to create an optimal user experience. Without this, health applications are at risk of not being widely adopted. This study uses the Design Thinking method to understand patient behavior and design relevant solutions. With stages such as empathy, problem definition, ideation, prototyping, and testing, this study aims to design a patient behavior model in health-based applications. This approach is expected to provide a comprehensive picture of the factors that influence patient behavior, as well as help developers create applications that are more intuitive, effective, and in accordance with user needs in the era of the pandemic and new habits
Analysis of Electronic Wallet User Sentiment on Twitter (x) Social Media Using the Naïve Bayes Classifier Algorithm
No full text Electronic wallets are one of the most popular payment methods in Indonesia with the number of users increasing significantly in recent years, including DANA, GoPay, and LinkAja. Along with the increasing number of users, the need to analyze user opinions and comments on social media, especially on Twitter (X) is also increasing. This study uses an experimental method with data collection from Twitter (X) using data crawling techniques. The dataset used is 1,500 data with 1 attribute. This study aims to analyze user sentiment towards electronic wallets using the Naive Bayes Classifier algorithm with the Python programming language. The results of the study showed that DANA had a negative sentiment of 16.6%, a neutral sentiment of 9.0%, and a positive sentiment of 74.4%. Followed by GoPay with a negative sentiment of 9.4%, a neutral sentiment of 11.4%, and a positive sentiment of 79.2%. Meanwhile, LinkAja had the lowest negative sentiment of 8%, with a neutral sentiment of 12.2% and a positive sentiment of 79.8%. The implementation of the Naive Bayes Classifier algorithm achieved an accuracy rate of 72% for DANA, 88% for GoPay, and 88%, for LinkAja
Perancangan Ulang Desain UI/UX Aplikasi I-Nusaplant Dengan Metode Design Thinking dan A/B Testing
No full textI-Nusaplant is a mobile-based application that can detect types of medicinal plants. The I-Nusaplant application was designed by Information Systems students using Android-based leaf images. This application aims to help the community in detecting medicinal plants. The ease of detecting medicinal plants using the application must also be supported by a good appearance and user experience. I-Nusaplant has 3 menus in it, namely Home, Detection, and About. The I-Nusaplant application has shortcomings after conducting interviews with medicinal plant experts, medicinal plant enthusiasts, and general users. The majority of respondents chose the I-Nusaplant application to be redesigned for several reasons related to user experience in running the application. Ease of obtaining information, time to move around each menu, and some features that users need. In doing a redesign, it is necessary to have an in-depth design using UI/UX. The design process in solving problems and generating user needs, researchers use the Design Thinking method to generate ideas to solve user problems. At the analysis stage, problem analysis is carried out, finding solutions to user problems, analyzing user needs. In the design stage, UI/UX design is produced, namely architecture information, user flow, and interface design. UI/UX design will be tested using the Maze tool. After that, compare the results of the old and new I-Nusaplant interface designs using the A/B Testing method. This method aims to see the performance of the old and new application designs. Our A/B testing revealed that the new design, while more complex, is just as efficient as the old one, both scoring 99 this shows that the new design is easy to use by users despite its different design
Klasterisasi Pola Penjualan Menu Makanan pada Rumah Makan menggunakan Metode K-Means Clustering
No full textThe culinary industry is one of the fastest-growing business sectors in Indonesia, as evidenced by the increasing number of restaurants emerging across the country. This intense competition demands that each restaurant develop effective strategies to attract customers and enhance profitability. One such strategy is analyzing menu sales patterns. This study contributes to the field of informatics, particularly in the application of data mining and machine learning techniques to support strategic decision-making in the culinary sector. The K-Means Clustering method was employed to analyze 12,404 daily sales transactions from a restaurant. The sales data were collected and analyzed to identify groups of menu items with similar sales characteristics. The research stages included data preparation, processing using RapidMiner and Microsoft Power BI, and analysis of the Clustering results. The quality of the clusters was evaluated using the Davies-Bouldin Index, which yielded a score of 0.354, indicating good separation and compactness between clusters. The analysis revealed that the optimal number of clusters is five, representing categories of highly popular, moderately popular, and less popular menu items. The most popular items include Chicken Rice, Tea, Catfish Rice, Chicken, and Potato Fritter. Meanwhile, the least preferred menu items include Minced Meat, Beef Tendon Rice, Jackfruit Curry, Beef Tendon, and Tempe. This Clustering provides valuable insights for restaurants to focus on developing popular menu items and consider improving or removing those that are less favored. The implementation of these Clustering results supports strategic decisions related to ingredient inventory management, menu promotion, and improvements in operational efficiency and customer satisfaction
Manfaat Blockchain pada Sistem Registrasi Tanah: Systematic Literature Review
No full textThe robust economic growth in Indonesia in the second quarter of 2023 indicates a projected population increase, leading to higher population density and driving the conversion of agricultural land. Land ownership has become a valuable source of capital, triggering intense competition. Land properties have emerged as valuable assets, emphasizing the importance of land registration as a process for recording ownership rights. The current centralized and manual land registration system faces challenges such as record duplications, unauthorized document reforms, and excessive departmental involvement. These weaknesses can result in pending cases, slow verification processes, and document manipulation.Digital transformation with blockchain technology is proposed as a solution for transparent, efficient, and legally certain land administration. This technology offers decentralized storage, resilience to changes, and peer-to-peer verification in transaction recording. While some countries have successfully implemented blockchain, others have faced failures due to environmental factors, state intervention, socio-political readiness, and institutional factors. In Indonesia, land conflicts have escalated, recording 562 cases from 1988 to July 2023. The lack of capacity and competence in local government human resources, coupled with suboptimal administration, complicates handling and hampers regional revenue. This research proposes a land registration framework with the implementation of blockchain as a solution to land administration issues in Indonesia
Prediksi Tinggi Gelombang Laut di Perairan Semarang – Demak dengan Menggunakan Random Forest dan XGBoost
No full textEnglish Climate change and regional oceanographic activities have contributed to increasing significant wave height (SWH) and sea level rise (SLR) along the northern coast of Central Java (Semarang-Demak). This study aims to analyze and predict SWH and SLR using two artificial intelligence methods: Random Forest (RF) and Extreme Gradient Boosting (XGBoost). The dataset includes meteorological and oceanographic parameters from 2019 to 2024. Model performance was evaluated using accuracy metrics such as RMSE, MAE, MAPE, and the coefficient of determination (R²). The results show that XGBoost outperforms RF in predicting both target variables. XGBoost achieved R² values of 0.9989 for SWH and 0.9921 for SLR, with MAPE scores of 1.6% and 1.1%, respectively. The most influential factor for SWH prediction was the historical significant wave height (hs), while the average daily sea level elevation had the highest impact on SLR prediction. Comparison plots between actual and predicted values indicate that both models effectively captured seasonal trends, particularly in identifying wave peaks in early months and sea level increases during mid-year.The 2025 forecast suggests rising SWH patterns from January to March and peak SLR values around June. These findings are expected to support coastal adaptation policies in response to climate change and to inform the design of more resilient marine infrastructure in the future
Sistem Presensi Otomatis Menggunakan Pengenalan Wajah Berbasis Deep Learning dan Real-Time Database
No full textThe attendance system is a crucial component in the operations of any organization. However, most existing attendance systems still require significant time or manual intervention from users. This study aims to develop a deep learning-based face recognition application with a real-time database to record attendance automatically. This approach is expected to make the attendance process more accurate, faster, and more convenient compared to traditional attendance methods. The study employs a quantitative method through primary data analysis from laboratory testing using dummy data. This testing aims to measure the accuracy of the face recognition system in automatically recording attendance. A face recognition application prototype has been successfully developed with real-time database integration using the Python programming language. The test results show that the application can recognize all faces in the database with a very high accuracy level. The system performance metrics indicate an accuracy of 99.1%, precision of 98.7%, recall of 98.7%, and F1-score of 98.7%. Additionally, the model has been implemented on an NVIDIA Jetson Nano mini-processor, demonstrating efficient operation on low-power hardware and real-time face recognition with optimal processing speed
CesLA (Cegah Stunting Lewat Anemia): Deteksi Anemia Non-Invasif pada Remaja Putri Berbasis Citra Konjungtiva
No full textStunting is a chronic nutritional problem that will directly affect the quality of human resources in the future. One of the contributing factors to stunting is anemia during pregnancy, which often originates from adolescence. Early detection of anemia in women of reproductive age is a crucial preventive measure to reduce the risk of stunting. This study aims to develop an anemia classification model based on conjunctival images using a combination of MobileNetV2 architecture and Support Vector Machine (SVM), and to implement the model into a mobile application named CeSLA (Cegah Stunting Lewat Anemia). The model was built using a dataset of female conjunctival images annotated based on haemoglobin levels and visual characteristics of the conjunctiva. Evaluation results explain that the model achieved precision, recall, and f1-score values ranging from 0.91 to 0.92 for each class, with a macro average of 0.92, indicating accurate and balanced classification performance. The trained and evaluated model was then integrated into the CeSLA mobile application. This application allows users, particularly adolescent girls, to detect potential anemia non-invasively by scanning the lower eyelid using a smartphone camera. CeSLA is also equipped with educational features such as health articles and a detection history log. With this approach, CeSLA is expected to serve as an innovative solution that supports early, self-administered anemia detection and contributes to the national effort to prevent stunting