Disaster management information system are essential for handling emergencies and saving lives through swift responses and coordination. This study aims to analyse and design an efficient backend technology for such apps. Research methods involve literature review, interviews, and observation with stakeholders. Load balancing is found effective for optimizing disaster management, efficiently distributing workloads and ensuring high system availability. The analysis identifies functional and non-functional requirements, examining technologies like API, WebSocket, Caching, and Relational Database. The design section creates a 3-tier architecture with caching for enhanced performance and scalability. Integration with Mobile Cognitive Radio Base Station (MCRBS) ensures emergency communication in affected areas.
Digital data sharing is an important part of information exchange in modern era. In rural areas this information exchange method is very limited. One method that can be used is offline sharing application that can be installed on smartphone. With this method, users can directly exchange data without the need for intermediaries or complicated setups. The widespread adoption rate of smartphones, data customization, and the convenience provided in this method compared to alternatives such as local area networks and physical hard drives. This situation brings a question as to 'Which offline data sharing methods implementations on Android can bring the most convenience and user experience in rural areas, for both data recipients and senders?’ This question guides the exploration of methodologies and technologies that can address the challenges of offline data sharing specifically on Android applications. Based on the comparisons that have been made, local hotspot was chosen as the main solution developed with Bluetooth as an alternative for devices that do not have the required features or hardware. The implementation applied to the application is by utilizing the local hotspot feature on Android provided by the WiFiManager class. The data content is then served on an android web server using the NanoHTTPD library. The data recipient can open the web server via IP address which can be accessed by connecting their device to the SSID of the sender device. The second alternative is to use Intent.SEND via the Bluetooth package to send data to other devices. Testing is carried out by sending normal data between one-to-one device for the two alternatives and one to many for the local hotspot alternative. Test results show that the implemented solution is successfully used to send JPG image with the size of 3.21MB and 657 pages PDF books with size of 20.8 MB.
In response to the critical shortage of skilled labor in Indonesia during the period from 2019 to 2021, this study conducts a comprehensive evaluation of the effectiveness of the Balai Besar Pelatihan Vokasi dan Produktivitas (BBPVP) through the application of advanced big data analysis techniques. Despite the government's successful efforts to achieve a 60% vocational and 40% academic student intake ratio by 2020, a persistent challenge remains: vocational education represents only 49% of total enrollments. The Kerangka Kualifikasi Nasional Indonesia (KKNI) framework plays a pivotal role in recognizing qualifications from diverse educational backgrounds. Nevertheless, ongoing difficulties persist in aligning the skillsets of vocational graduates with the evolving demands of various industries, resulting in elevated unemployment rates. This research underscores the urgency for continuous improvements in vocational education and the establishment of more robust partnerships between vocational institutions and industries. This collaborative approach aims to equip graduates with the necessary skillsets demanded by employers, thus enhancing their employability and reducing unemployment. By aligning BBPVP's strategies with the dynamic requirements of the labor market and fostering adaptability to industry changes, Indonesia can nurture a highly proficient workforce, ultimately driving economic growth and prosperity. The study underscores the vital need for strategic collaborations between educational institutions, governmental bodies, and industry stakeholders to fully unlock the potential of vocational education within the country. BBPVP must remain steadfast in evolving its training programs to address industry-specific needs while concurrently promoting the societal value of vocational education. Raising awareness about career opportunities and the advantages of vocational training will incentivize more students to pursue vocational pathways, consequently strengthening the talent pool available to various industries. In conclusion, this study underscores the pivotal role played by BBPVP in shaping the nation's workforce and its substantial contribution to Indonesia's economic growth and development.
Promotion abuse fraud is promotion abuse by duplicating accounts to gain an advantage over promotional codes fraudulently. This action inflicts a financial loss on the company. Therefore, this study aims to deal with fraud by developing an application to detect promotion abuse fraud. The application development process includes needs analysis, modeling, and application development. The dataset used for modeling comes from an e-commerce company in Indonesia. The dataset collection stage includes calculating the similarity between accounts using the Levenshtein distance similarity algorithm to get additional features, such as the number of accounts that are similar to an account. At the modeling stage, experiments were carried out using the Random Forest algorithm and a risk scoring algorithm based on machine learning, namely FasterRisk. The FasterRisk has better performance than the Random Forest algorithm shown by a higher F1 score and AUC score. The algorithm also has an advantage in terms of interpretability because it has an output in the form of a more understandable risk score model, so users can understand the factors that are fraud indicators. The FasterRisk prediction result is then deployed into a web application based on user needs.
The stealthy Mycobacterium tuberculosis breeds an infectious scourge known as Tuberculosis (TB). It is a bacterial infection related chronic lung disease. An early and accurate diagnosis of TB is essential. chest X-ray images are used to detect this disease. However, manual detection takes longer, which occasionally leads to errors. In this article, we proposed a deep learning approach for TB classification using chest X-rays. Some deep learning algorithms search through a hypothesis space in search of an appropriate hypothesis that will produce accurate predictions for a certain disease scenario. To detect tuberculosis in chest X-rays, we use Res Net a deep learning architecture recognized for its capacity to make use of feature reuse and overcome vanishing gradient issues. In order to provide a more exact hypothesis for the prediction of tuberculosis using the Res Net, a deep learning technique, this proposed model is developed on top of prior research. We were able to reach 99% accuracy with the classifier that was based on deep learning.
