Home collection (NIMS)
URI permanente para esta coleção:
Navegar
Entradas recentes
- Prototype Retrieval-Augmented Federated Learning System for Robust Intrusion DetectionPublication . Zhou, Hanlin; Yan, Huiru; Nian, Jiawei; Liu, Cong; Wang, Ying; Chen, Xiaomin; Theodoropoulo, Georgios; Cheng, Long; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; IEEE PressDetecting malicious attacks is essential for protecting computer systems and ensuring device security. Federated Learning (FL)-based Intrusion Detection Systems (IDS) have emerged as promising solutions, enabling multiple clients (i.e., data owners) to collaboratively train intrusion detection models without sharing private data. However, current FL studies typically assume that each client’s training and test label distributions are identical. This assumption is overly idealistic and rarely holds in real-world scenarios, leading to suboptimal performance when label distribution shifts occur between the training and testing data. To address this challenge, we propose FedPRO, a plug-and-play framework designed to improve the test-time performance of existing FL methods, without modifying their original training pipelines or fine-tuning the trained FL models. Specifically, we develop a unique prototype generation and optimization mechanism to produce semantically meaningful class prototypes. These prototypes constitute a prototype memory bank, serving as an external knowledge repository. At test time, a prototype retrieval-augmented inference strategy is employed to query relevant prototypes and refine predictions on each client, effectively alleviating the label distribution shift issues and boosting prediction accuracy. We evaluate FedPRO by integrating it with various off-the-shelf FL methods. Extensive experimental results on benchmark datasets demonstrate its effectiveness and scalability. Notably, applying FedPRO to the state-of-the-art method FedDBE improves its test accuracy from 79.25% to86.66% on the CICIDS-2018 dataset, while introducing only approximately 32KB of additional communication overhead. Our code is available at: https://github.com/zza234s/FedPRO
- The Growth of Digital Transformation in European CompaniesPublication . Santos, Victor; Malta, Pedro; NOVA Information Management School (NOVA IMS)The accelerated digital transformation of European companies reveals a heterogeneous landscape in the adoption of Artificial Intelligence (AI). While many organizations primarily employ AI for analytical purposes such as data mining, forecasting, or customer insights, a growing number of countries are integrating AI not only into analytics but also into decision-making processes. This dual application marks a critical shift, as firms leverage predictive algorithms and real-time systems to support marketing strategies, e-commerce growth, and broader business management. Drawing on Eurostat datasets, this paper maps the adoption of AI across EU member states, distinguishing between analytical and decision-making uses. By combining state-of-the-art literature with a predictive analysis of adoption patterns, we demonstrate how countries that embed AI in decision-making achieve greater alignment between data-driven intelligence and organizational strategy. Our findings highlight emerging leaders in AI-enabled decision processes in Europe and propose a research agenda on how AI can enhance competitiveness, innovation, and value creation in marketing, e-commerce, and business ecosystems.