Aplikasi perpesanan sederhana lintas platform yang memungkinkan untuk bertukar pesan, mulai dari pesan teks, gambar, video, serta panggilan video dan panggilan cepat tanpa biaya banyak digunakan oleh pengguna di Indonesia, salah satunya adalah Aplikasi Whatsapp. Platform tersebut banyak digunakan dan diminati semua kalangan, dalam percakapan pribadi maupun percakapan grup. Percakapan yang di dominasi dengan menggunakan pesan teks sangat memungkinkan terjadinya kesalahan pengetikan dalam pengiriman pesan yang mengakibatkan kesalahan persepsi pembaca pesan tersebut. Berdasarkan kesalahan pengiriman pesan dan kesalahan dalam pengetikan pesan teks pengguna platform tersebut maka akan dilakukan analisa pengguna aplikasi Whatsapp di lingkungan Universitas XYZ menggunakan metode klasifikasi dengan algoritma Naïve Bayes dan KNN terhadap kesalahan mengirim pesan teks. Dari data yang di dapat yaitu hasil klasifikasi menggunakan algoritma Naïve Bayes mendapatkan akurasi sebesar 75%, sedangkan klasifikasi menggunakan algoritma KNN mendapatkan akurasi sebesar 66,7%. Dari hasil klasifikasi tersebut performa algoritma Naïve Bayes lebih besar dari pada performa algoritma KNN.
The COBIT 5 framework can be implemented in all organizations or enterprises. This study analyzes and finds the maturity level of IT governance in the XYZ Library. Libraries have a role in encouraging the efficiency and effectiveness of the learning process. The IT governance element in COBIT 5 aims to get results from evaluating stakeholders' needs, conditions and choices. Survey and in-person interviews are methods of obtaining data and information from the IT governance process at the XYZ Library. The COBIT 5 domains used are evaluated, direct, and monitored (EDM) and align, plan and organize (APO) domains. The results of the evaluation that have been carried out show the findings of gaps in the process domain EDM01 with a value of 2.34 and EDM04 with a gap of 2.25. While in the process domain, APO01 with a gap of 2.13, and APO07 with a gap of 2.34. With the average value of the entire gap at the level of 2.26, the findings of the XYZ Library with the process domains EDM01, EDM04, APO01, and APO07 show that the process is managed and the results are determined, controlled, and maintained (Managed). For further research, we can add more process domains both in the Governance and Management areas so that the audit process with the COBIT 5 framework can be carried out comprehensively.
Flood disasters that are monitored in real-time on social media can be seen to report directly the condition of the affected areas. Areas that have been warned to be affected by the floods were informed via social media. Surrounding areas that are likely to be affected can be more vigilant by directly speeding up information and getting responses from social media users who are around flood-prone areas. The research aims to provide visualization models that can accelerate flood response information from disaster management, track disasters with increasing vigilance, and accelerate flood disaster recovery with analysis on social media. The approach used with Natural Language Processing (NLP), a data source derived from posts on Instagram is taken for analytical materials. Data sources from Instagram with flood hashtags in Kalimantan are used by using the Natural Language Processing (NLP) process stage to get core information visualizations to speed up flood response information. Visualization of social media data information used based on extracting information from Instagram posts, responses, and hashtags to speed up information provides troubleshooting and the importance of speeding up flood response information. The results of data visualization can accelerate disaster response information to increase awareness of the condition of the surrounding area that can be affected by floods, it can be seen that the amount of data on the hashtag provides data visualization information in accelerating flood disaster management from disaster tracking to disaster recovery.
Asphyxia is influenced by several factors, including the factors affecting the Immediate Was maternal factors That relates Conditions mother Pregnancy and childbirth such as hypoxia mother, Asphyxia factor data can be modeled using the classification approach. this paper will be compared k-nearest neighbor algorithm and Naive Bayes classifier to classify asphyxia factor. Naive Bayes uses the concept of Bayes’ Theorem which assuming the independency between predictors. Basically, Bayes theorem is used to compute the subsequent probabilities. Analysis of the two algorithms has been done on several parameters such as Kappa statistics, classification error, precision, recall, F-measure and AUC. We achieved the best classification accuracy with KNN algorithm, 92,27%, for k=4. are lower than the rates achieved with Naïve Bayes 83,19%.
Ilmu pengetahuan dan teknologi telah mengubah kehidupan manusia saat ini, hampir semua sektor industri dibantu oleh mesin kecerdasan buatan, revolusi industri generasi keempat membawa kehidupan manusia selalu berdampingan dengan sistem cerdas, selain sektor industri, pertanian saat ini sedang mengalami revolusi digital untuk dapat memberdayakan petani, salah satunya adalah mengoptimalkan penggunaan irigasi sebagai ketentuan dan pengaturan air untuk mendukung pertanian. Penelitian ini bertujuan untuk membuat prototipe irigasi digital berbasis mikrokontroler yang dapat memberdayakan petani dan menyediakan sistem pendukung keputusan bagi petani dalam menghadapi krisis air di lahan kering. Dengan menggunakan sistem berbasis pengontrol, petani akan dapat memantau dan mengontrol sistem pasokan air menggunakan smartphone yang terhubung ke sistem dan dapat membantu petani dalam membuat keputusan untuk tanah mereka. Dengan memanfaatkan teknologi Internet of Things (IoT) berbasis mikrokontroler menggunakan raspberry Pi dan Arduino, di mana prototipe irigasi digital dijalankan menggunakan smartphone yang mengirimkan data ke server berbasis mikrokontroler dan diterima oleh sensor yang telah terhubung menggunakan raspberry dan Arduino.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.