Hotel booking cancellations pose significant challenges to the hospitality industry, affecting revenue management, demand forecasting, and operational efficiency. This study explores the application of machine learning techniques to predict hotel booking cancellations, leveraging structured data derived from hotel management systems. Various classification algorithms, including Random Forest, XGBoost, and LightGBM were evaluated to identify the most effective predictive model. The findings reveal that XGBoost model outperforms other models, achieving F2-score of 0.7897. Key influencing factors include deposit type, total number of special requests, and marketing segment. The results underscore the potential of predictive modeling in optimizing hotel revenue strategies by enabling proactive measures such as dynamic pricing, targeted customer engagement, and improved overbooking policies. This study contributes to the ongoing advancements in data-driven decision-making within the hospitality industry, offering insights into how machine learning can mitigate financial risks associated with booking cancellations.

The IoT (Internet of Things) and Digital Transformation - Skill For All (SFA) training for vocational school teachers in Kulon Progo is an information technology training program. The SFA training addresses the skills gap by offering knowledge and skills necessary to work with IoT devices using IoT Kits from the Network Technology and Application lab, which is evolving in the digital transformation era. This training provides a solid foundation for managing IoT devices, which are highly relevant to today's industry needs. Configuring, managing, and programming IoT devices is essential in the digital age. By participating in this training, vocational school teachers in Kulon Progo can acquire practical skills directly applicable to their curriculum, helping students understand IoT concepts more deeply and practically. The IoT (Internet of Things) and Digital Transformation - Skill For All (SFA) training is designed to provide a comprehensive understanding of networking basics and how to configure and manage IoT Kits devices

Dairy cattle are ruminant livestock that produce milk whose productivity is very dependent on maintenance management. This community service activity aims to increase the knowledge and skills of dairy farmers in Raraa Village, East Kolaka Regency. This activity was carried out in several stages, including: 1) Socialization of dairy cattle development; 2. Demonstration on the maintenance and feeding of dairy cows; 3. Monitoring and Evaluation. During the socialization activities, the breeders were very enthusiastic in listening and active in the discussion because the breeders in Raraa Village were just getting to know and starting to raise dairy cows. Farmers are also very focused on paying attention to explanations and technical guidance on the maintenance, management and feeding of dairy cows during direct visits to the breeder's pen. At the end of the activity, an evaluation was carried out to determine the level of understanding of dairy farmers. Monitoring and assistance to farmers is also carried out to help farmers who experience difficulties in the process of caring for and feeding dairy cows. This activity can increase the knowledge and skills of dairy farmers so that farmers can maintain and develop dairy cattle in accordance with the government's goal of distributing dairy cattle seeds in the Southeast Sulawesi region

In recent years, technological developments and paradigm shifts in the world of education and research demand a transformation in the management system of research and community service. The Institute for Research and Community Service (LPPM) of Universitas Negeri Semarang (UNNES) in its role of realising the responsibility of the Tri Dharma of Higher Education, has made a breakthrough in the field of research and community service through an application called SIPP (Research and Community Service Management Information System) which can be accessed on the website. However, there are still five obstacles that need to be improved. This research intends to identify concrete problems faced by Universitas Negeri Semarang in this regard, and design solutions that can increase the effectiveness of the research and community service management system. The results showed that the development of an Artificial Intelligence (AI)-based Recommendation System has become one of the most effective strategies in enhancing user experience in improving the effectiveness of the research and community service management system at Semarang State University. By utilising AI technology, the SIPP application can optimise the data management process, identify research trends, and facilitate collaboration between researchers. 

Pemahaman konsep mahasiswa di perguruan tinggi dalam memahami materi gelombang bunyi merupakan salah satu aspek krusial dalam studi fisika. Dengan pemahaman konsep dasar yang berbeda-beda, dapat menyebabkan kesalahan konsep. Untuk mengetahui pemahaman konsep, maka dari itu dilakukan pengembangan instrumen tes diagnostik yang akurat dan efektif. Penelitian ini bertujuan untuk (1) mengetahui karakteristik instrumen tes yang dikembangkan (2) mengetahui kualitas instrumes tes yang dikembangkan (3) mengidentifikasi pemahaman konsep pada materi gelombang bunyi. Jenis penelitian adalah Research & Development (R&D) model 4D dengan pengumpulan data melalui kuisioner dan angket yang diberikan kepada mahasiswa fisika dasar 2. Instrumen tes yang dikembangkan berjumlah 15 soal yang terdiri dari pilihan ganda tiga tingkat. Hasil uji validitas ahli terhadap instrumen tes diagnostik yang dikembangkan sebesar 89,21% sehingga memiliki kriteria sangat layak. Hasil uji skala kecil menunjukkan bahwa instrumen tes diagnostik yang dikembangkan terdapat 13 valid dan reliabel. Tingkat kesukaran soal menunjukkan 2 soal masuk kategori mudah, 12 soal masuk kategori sedang, dan 1 soal masuk kategori sukar. Hasil daya pembeda dari uji skala kecil menunjukkan 2 soal masuk kategori buruk, 5 soal masuk kategori cukup, dan 8 soal masuk kategori baik. Identifikasi pemahaman konsep dari uji skala kecil menunjukkan hasil rata-rata tingkat pemahaman konsep 32,32%, tidak paham konsep 39,65%, dan mengalami miskonsepsi 27,02%. Mahasiswa paham konsep tertinggi pada indikator ke-1 sebesar 69,70% dengan sub konsep intensitas dan taraf intensitas bunyi serta mahasiswa tidak paham konsep tertinggi pada indikator ke-13 dan 14 sebesar 66,67% pada sub konsep intensitas dan taraf intensitas bunyi. Dilihat dari proses kognitif memahami, indikator mencontohkan memiliki pemahaman konsep lebih baik dengan kriteria sedang dan indikator membandingkan memiliki pemahaman konsep dengan kriteria sangat rendah. 

Pada sebuah departemen di PT. X terdapat reaktor batch berkapasitas 300 L yang berfungsi utama untuk mensintesis produk berbasis reaksi. Saat ini reaktor tersebut belum dilengkapi jalur pendingin aktif, sehingga kontrol suhu kurang optimal dan proses pendinginan pasca-reaksi berlangsung sangat lambat. Penelitian ini bertujuan merancang penambahan jalur koil pendingin dan menghitung kebutuhan kapasitas pendinginannya agar suhu produk dapat diturunkan dari 110?°C menjadi 60?°C secara cepat dan stabil, serta mengevaluasi kinerja termal sistem setelah modifikasi. Analisis beban panas dilakukan dengan persamaan Q = m•Cp•?T (m=300 kg, Cp=4180 J/kg•K, ?T=50?K) menghasilkan Q ? 62,7 MJ. Sistem pendingin baru berupa koil 10 lilitan terbuat dari Stainless Steel 316 dengan diameter pipa 1 inchi di sekeliling reaktor dirancang untuk mendapatkan laju aliran air pendingin 3,08 kg/s (11,08 m³/jam). Perhitungan menunjukkan kapasitas pendinginan baru sekitar 193 kW dengan selisih suhu air 15?K, dan prediksi waktu pendinginan hanya ~5,4 menit. Hasil pengujian menampilkan suhu akhir produk 60?°C dalam 5,4 menit (97% lebih cepat dari 3 jam sebelumnya). Penambahan jalur pendingin terbukti menurunkan viskositas produk dari 8.820 cP menjadi 6.980 cP (dalam rentang spesifikasi 2.000–8.000 cP) dan meningkatkan kandungan padatan dari 38,62% menjadi 39,51. Hal ini menunjukkan sistem pendingin baru berhasil mengoptimalkan sifat reologi dan kualitas produk melalui kontrol suhu yang lebih presisi.

Customer churn is a major challenge in e-commerce, directly affecting revenue and profit. This study aims to develop a machine learning model using XGBoost to predict churn probability. To handle class imbalance, SMOTE was applied as a resampling method, and hyperparameter tuning was performed to enhance performance. The model was evaluated using the F2-score, prioritizing recall while maintaining precision. The results show that the XGBoost model with SMOTE achieves strong performance, with an F2-score of 0.849 on the tuned test data. This model can help businesses identify at-risk customers early, enabling proactive retention strategies.

This study aims to develop an intuitive and efficient smart home control system by utilizing hand tracking and speech recognition technologies. These technologies employ the OpenCV, Mediapipe, PyAudio, and Speech Recognition libraries to recognize hand gestures and voice commands in real-time. The system is developed using a Raspberry Pi connected to a webcam and microphone as input devices, and a relay to control electronic appliances. The results show a high accuracy rate at optimal light intensity for hand tracking and a specific distance for speech recognition. This system is implemented in an IoT environment to control devices such as lights and door locks. The research is expected to contribute to the development of smarter and more user-friendly smart homes.

Gangguan gastrointestinal fungsional (Functional Gastrointestinal Disorders/FGIDs) adalah kondisi kronis yang ditandai dengan gejala yang signifikan dan mempengaruhi kualitas hidup penderitanya. Penelitian ini mengkaji peran dukungan sosial dalam mengelola gejala FGIDs melalui narrative review yang komprehensif. Temuan menunjukkan bahwa dukungan sosial, termasuk dukungan emosional, instrumental, dan informasi, memiliki dampak positif yang signifikan terhadap kesejahteraan fisik dan mental individu dengan FGIDs. Dukungan sosial membantu mengurangi stres dan kecemasan, yang sering memperburuk gejala FGIDs, dengan menurunkan aktivitas sistem saraf simpatis dan meningkatkan aktivitas saraf vagus. Efektivitas dukungan sosial dipengaruhi oleh faktor budaya, kondisi ekonomi, dan stabilitas hubungan sosial. Penelitian juga menyoroti peran penting teknologi dan media sosial dalam menyediakan dukungan sosial yang lebih luas dan mudah diakses, meskipun penting untuk memastikan informasi yang diberikan dapat dipercaya dan berbasis bukti. Implikasi untuk praktik klinis mencakup integrasi dukungan sosial dalam rencana perawatan dan edukasi pasien serta keluarga. Penelitian lebih lanjut diperlukan untuk mengeksplorasi mekanisme spesifik dukungan sosial dan mengembangkan intervensi inovatif. Dengan memahami dan memanfaatkan kekuatan dukungan sosial, kita dapat membantu individu dengan FGIDs mengelola kondisi mereka dengan lebih baik dan meningkatkan kualitas hidup mereka.

Settlement is a basic human need that is closely related to the feasibility of life and well-being. The purpose of this study is to gain a better understanding of settlement patterns that occur in rural and urban areas, and to investigate the factors that influence these patterns. This study collects, analyzes, and synthesizes related literature. The results show that settlement patterns in urban areas are denser and develop vertically, while settlement patterns in rural areas tend to be spread out and horizontal. Various factors, such as natural conditions (such as climate, topography, water and land availability), socio-economic factors (such as employment, education, and income), and government policies and local cultural developments, influence these differences. These results emphasize that understanding settlement patterns as a whole is essential to building areas that are equitable, sustainable, and responsive to community needs.