Published in ITM Web of Conferences
This paper explores the potential of Twitter, a popular social media platform, as a tool for predicting election outcomes. Sentiment analysis has emerged as a powerful tool for predicting election outcomes, with numerous studies showcasing its effectiveness in various countries. For instance, research has utilized sentiment analysis to forecast ele...
Published in ITM Web of Conferences
Given the current situation of the economy, credit card use has increased significantly. Users can make significant cash payments with these cards without carrying a lot of cash on them. They have simplified the process of conducting cashless transactions and enabled consumers to make payments of any kind with greater ease. While there are many ben...
Ce rapport détaille les approches testées dans le cadre de l’atelier EvalLLM 2024. Notre participation à l’atelier a eu pour intérêt d’apporter des éléments de réponses aux questions suivantes : dans quelle mesure les données annotées et développées dans le cadre du projet POPCORN peuvent-elles s’adapter à une nouvelle typologie de classes et à un ...
Práce zdůrazňuje efektivitu metod strojového učení v různých vědeckých oblastech a zdůrazňuje cíl vývoje přesných matematických modelů pro složité systémy. Je nastíněna konkrétní aplikace zaměřená na tvorbu digitálních dvojčat pro mechanické struktury. Navržený přístup hybridního modelování představuje novou integraci metody konečných prvků (FEM) a...
L'abstract è presente nell'allegato / the abstract is in the attachment
There has been an increasing interest in fraud detection methods, driven by new regulations and by the financial losses linked to fraud. One of the state-of-the-art methods to fight fraud is network analytics. Network analytics leverages the interactions between different entities to detect complex patterns that are indicative of fraud. However, ne...
The empirical risk minimization (ERM) problem with relative entropy regularization (ERM-RER) is investigated under the assumption that the reference measure is a $\sigma$-finite measure, and not necessarily a probability measure. Under this assumption, which leads to a generalization of the ERM-RER problem allowing a larger degree of flexibility fo...
Machine learning models always make a prediction, even when it is likely to be inaccurate. This behavior should be avoided in many decision support applications, where mistakes can have severe consequences. Albeit already studied in 1970, machine learning with rejection recently gained interest. This machine learning subfield enables machine learni...
Published in E3S Web of Conferences
The development of machine learning (ML) methods in the field of material science has provided new possibilities for predictive modeling, especially in the field of mechanical material evaluation. The study provides an in-depth investigation of the utilization of various machine learning methods in predicting of mechanical characteristics throughou...
We consider the neural ODE and optimal control perspective of supervised learning, with $\ell^1$-control penalties, where rather than only minimizing a final cost (the empirical risk) for the state, we integrate this cost over the entire time horizon. We prove that any optimal control (for this cost) vanishes beyond some positive stopping time. Whe...
