The ongoing COVID-19 pandemic has caused worldwide socioeconomic unrest, forcing governments to introduce extreme measures reduce its spread. Being able accurately forecast when the outbreak will hit peak would significantly diminish impact of disease, as it allow alter their policy accordingly and plan ahead for preventive steps needed such public health messaging, raising awareness citizens increasing capacity system. This study investigated accuracy a variety time series modeling...

Abstract A large variety of issues influence the success data mining on a given problem. Two primary and important are representation quality dataset. Specifically, if much redundant unrelated or noisy unreliable information is presented, then knowledge discovery becomes very difficult It well-known that preparation steps require significant processing time in machine learning tasks. would be helpful quite useful there were various preprocessing algorithms with same reliable effective...

Smart cities, leveraging advanced data analytics, predictive models, and digital twin techniques, offer a transformative model for sustainable urban development. Predictive analytics is critical to proactive planning, enabling cities adapt evolving challenges. Concurrently, techniques provide virtual replica of the environment, fostering real-time monitoring, simulation, analysis systems. This study underscores significance systems support test scenarios that identify bottlenecks enhance...

The ability to predict a student's performance could be useful in great number of different ways associated with university-level distance learning. Students' key demographic characteristics and their marks on few written assignments can constitute the training set for supervised machine learning algorithm. algorithm then able new students, thus becoming tool identifying predicted poor performers. scope this work is compare some state art algorithms. Two experiments have been conducted six...

This paper explores the effectiveness of machine learning techniques in detecting firms that issue fraudulent financial statements (FFS) and deals with identification factors associated to FFS. To this end, a number experiments have been conducted using representative algorithms, which were trained data set 164 fraud non-fraud Greek recent period 2001-2002. The decision particular method choose is complicated problem. A good alternative choosing only one create hybrid forecasting system...

Transferring knowledge from one domain to another has gained a lot of attention among scientists in recent years. Transfer learning is machine approach aiming exploit the retrieved problem for improving predictive performance model different but related problem. This particularly case when there lack data regarding problem, plenty about one. To this end, present study intends investigate effectiveness transfer deep neural networks task students’ prediction higher education. Since building...

Educational Data Mining (EDM) has emerged over the last two decades, concerning with development and implementation of data mining methods in order to facilitate analysis vast amounts originating from a wide variety educational contexts. Predicting students’ progression learning outcomes, such as dropout, performance course grades, is regarded among most important tasks EDM field. Therefore, applying appropriate machine algorithms for building accurate predictive models outmost importance...

Bagging and boosting are among the most popular re- sampling ensemble methods that generate combine a diversity of classifiers using same learning algorithm for base-classifiers. Boosting algorithms considered stronger than bagging on noise- free data. However, there strong empirical indications is much more robust in noisy settings. For this reason, work we built an voting methodology ensembles with 10 sub- each one. We performed comparison simple 25 sub-classifiers, as well other known...