Publication Search

Android continues to innovate, and biometric systems particularly fingerprint sensors must operate with high efficiency. However, the main challenge lies in the communication between hardware and software, which often causes high latency, excessive power consumption, and protocol incompatibility between components. As a result, authentication speed decreases and system stability is compromised, especially on mid- to low-range devices. This study explores various implementations of Android-based fingerprint systems, focusing on how sensor modules interact with the microcontroller or Trusted Execution Environment (TEE) and the operating system through a hardware-software co-design approach to evaluate integration efficiency across all layers. The analysis reveals that conventional protocols such as Bluetooth or serial connections still cause delays, while improvements in drivers and the Hardware Abstraction Layer (HAL) can significantly reduce latency. As a solution, the researchers propose a co-design optimization approach that utilizes data flow normalization within the HAL and adopts lightweight communication protocols to accelerate the verification process. Based on the test results, this approach successfully improves efficiency—authentication time is reduced by up to 35% and power consumption decreases by approximately 15%. Therefore, the efficiency of communication between hardware and software becomes a key factor in enhancing the performance and reliability of fingerprint systems on Android devices.

Number recognition is a critical skill in early education, laying the foundation for digital literacy and mathematical understanding. However, traditional methods of teaching number recognition often lack the interactivity and engagement necessary for effective learning, particularly for young learners. This study proposes the development of a deep learning-based Android application aimed at enhancing the number recognition process by providing an interactive and visual learning experience. The application utilizes a Convolutional Neural Network (CNN), a type of deep learning model, to recognize handwritten numbers, offering users an innovative and engaging way to learn and practice number recognition.In the proposed application, users can draw numbers on their devices, and the system will immediately provide feedback regarding the accuracy of the drawn numbers. The CNN model will be trained on a comprehensive dataset of handwritten digits to ensure high accuracy in recognition. This real-time feedback loop is designed to help users learn the correct form of numbers while also introducing the foundational concepts of deep learning. The main objective of this application is to provide an accessible, interactive platform for learning number recognition, especially for novice learners who may be unfamiliar with basic concepts in machine learning and digital literacy. By integrating deep learning technology, the application not only supports the learning of number recognition but also serves as an introduction to artificial intelligence (AI) concepts in a practical, easy-to-understand format. Furthermore, the application is designed to be user-friendly, ensuring that it is suitable for a wide range of learners, including children and beginners. The application aims to combine the fundamental principles of deep learning with a practical, hands-on learning experience, fostering a deeper understanding of both number concepts and the potential of AI in everyday life.

A wedding is an important event in a person's life that requires various careful preparations, one of which is sending invitations to guests. This research aims to create an Android-based wedding invitation booking application that utilizes Augmented Reality (AR) technology with a Design Thinking approach. This approach includes five main stages, namely empathize, define, ideate, prototype, and testing, to ensure that the developed application is truly in accordance with the needs and expectations of users. With AR technology, users can view and interact directly with virtual invitations through digital devices such as smartphones, which provides a more engaging, interactive, and modern visual experience than conventional physical invitations. The app's key features include invitation design selection, content customization (name, location, date, and couple's photo), as well as simulated invitation display with three-dimensional effects via AR. The app also allows users to store, share, and place orders directly through an integrated system. The results of the evaluation conducted using the System Usability Scale (SUS) method showed a score of 75.3. This score indicates that the app has a good usability level and is positively received by the majority of respondents. In addition, this application is considered to be able to increase efficiency in the invitation ordering process and provide a more modern and personal impression to users and invited guests. The conclusion of this study is that AR-based wedding invitation booking apps not only serve as a practical tool, but also become digital innovations that enrich the user experience in modern weddings. This research is expected to make a real contribution to the development of creative information technology and inspire other developers to create similar solutions that are innovative, efficient, and relevant to the needs of the times.

Mental health is a crucial aspect of modern life, with stress and anxiety being among the most common and impactful psychological disorders. This research proposes a stress and anxiety monitoring system based on the Internet of Things (IoT), integrating biometric sensors and Deep Neural Networks (DNN) for early detection and in-depth analysis. The system is designed using MAX30102 (heart rate and SpO₂), GSR (Galvanic Skin Response), and DS18B20 (body temperature) sensors, managed by an ESP32 microcontroller and communicating through the MQTT protocol. Physiological data is collected in real-time, formatted in JSON, and transmitted to both Android and web-based applications for visualization. The DNN model is developed using the TensorFlow framework with a layered architecture and ReLU activation functions to classify four mental states: relaxed, calm, anxious, and highly stressed. The training dataset comprises both primary and secondary data, including the WESAD dataset. Model performance is evaluated through k-fold cross-validation, showing high accuracy and strong generalization capabilities. The results indicate that the integration of sensor technology and deep learning significantly improves the effectiveness of stress and anxiety detection compared to traditional methods. This system demonstrates great potential for the development of AI-based wearable devices for autonomous, real-time, and adaptive mental health monitoring. 

This research presents a personalized, context-aware recommender system to suggest Places of Interest (POIs) using a hybrid approach combining Bayesian inference and collaborative filtering. The system explicitly addresses the cold-start problem that new users face and improves recommendation accuracy by considering contextual variables such as user mood, budget, companion, and location. The system collects real-time contextual inputs for new users with no historical data and applies Bayesian inference to generate relevant POI suggestions. As users begin to interact and provide ratings, the system progressively shifts to a collaborative filtering mechanism, leveraging cosine similarity to identify similar users within comparable contexts. The recommender system focuses on three categories of POIs: restaurants, hotels, and landmarks. These locations are retrieved through the Google Maps API, and only mapped locations are considered. The system was implemented on Android devices and evaluated through a user study involving 25 participants from diverse backgrounds, including software developers, IT students, and general users. Evaluation metrics such as normalized Discounted Cumulative Gain (nDCG) and classification accuracy were used to assess recommendation quality. Results demonstrate that the system performs better than traditional methods, with nDCG improvements reaching up to 83 percent. Users reported high satisfaction regarding the recommendations' accuracy, ease of use, and contextual relevance. While the system offers significant improvements, it also has certain limitations. Its dependency on Google Maps data may restrict its scope, and using only four contextual factors limits the system’s adaptability to more complex user preferences. Future enhancements could include additional dynamic contexts such as weather, POI popularity, and time-related trends, as well as integrating more advanced models to increase personalization and flexibility in real-world applications.

This study aims to develop a Learning Management System (LMS) for kepalangmerahan management education in the Indonesian Red Cross Youth at SMK Wiworotomo Purwokerto The methodology employed is the Knowledge Management System Life Cycle (KMSLC) with the Meyer and Zack cycle approach, along with the spiral knowledge model by Nonaka and Takeuchi. Network infrastructure involves the internet, wireless access points, and private internet networks. Hardware and software include android devices with browser applications. Resource and data identification is carried out through discussions, interviews, and data entry, utilizing sources from PMR trainers, the Indonesian Red Cross website, and scholarly literature. The Knowledge Management (KM) team consists of PMR trainers, LMS developers, administrators, and PMR members. The system is implemented using WordPress and the TutorLMS plugin, with verification and validation testing conducted through black box testing. The research results demonstrate the successful development of the kepalangmerahan management education LMS in PMR SMK Wiworotomo Purwokerto, with a specific focus on disaster management education materials. This study contributes to improving effective learning access within PMR.

The introduction of hijaiyah letters in early childhood is an important stage in building the basis for learning the Qur'an. This research aims to develop an interactive game called Al-Dhaki which is designed to help parents and early childhood education teachers in introducing hijaiyah letters to children in an interesting and effective way. This game is designed using quantitative research methods to analyze user needs, with a 5W+1H approach (What, Why, Who, When, Where, and How) in data collection and analysis. The results of the study show that game-based interactive learning media can increase children's interest in learning and simplify the process of recognizing hijaiyah letters. Al-Dhaki is developed in the form of an application file (APK) that can be installed on Android devices, making it easily accessible to users. The game is equipped with interactive features, such as sounds, animations, and simple quizzes, to increase children's engagement in learning. With the presence of Al-Dhaki, it is hoped that the process of recognizing hijaiyah letters will be more fun and effective, as well as providing practical solutions for parents and PAUD teachers in supporting the development of religious literacy in early childhood.