EBTIC

Federated Learning (FL) enables decentralized model training without sharing raw data, but it remains vulnerable to various privacy threats. Cryptographic methods offer protection but often introduce high computational and communication overhead, degrading performance and reducing parameter utility. In this paper, we present Hash-comb, a lightweight, privacy-preserving, distance-aware hashing approach for privatizing model updates in FL. We detail the algorithm’s design, its integration into FL workflows, and its encoding/decoding processes. We begin by presenting background on privacy threats in FL, followed by a literature review of cryptographic approaches used to protect privacy in FL systems. We then introduce the Hash-comb approach, explaining its core principles and how it encodes data using distance-aware, multi-level quantization. This is followed by a detailed methodology outlining the encoding and decoding processes, as well as how Hash-comb integrates into the FL workflow. In our experiments, we evaluate Hash-comb across multiple datasets, Machine Learning (ML) models, and data distribution scenarios. The results demonstrate that Hash-comb maintains global model accuracy comparable to that of a conventional FL setup, even after encoding and decoding. These findings confirm that Hash-comb guarantees privacy with minimal impact on model utility, making it a practical and scalable solution for secure and privacy-preserving FL.

Chess is a complex logical game involving ongoing strategic forward planning and evaluation. Solving chess puzzles is one of the most common ways of training and developing chess skills. It involves continuing the game from a certain initial chessboard state against a real or AI opponent until defeat or a significant advantage is achieved. To ensure solving chess puzzles is efficient and engaging for the real player, it is important to understand the difficulty of the puzzle and match it with the skill of the solver. Accurate and fast assessment of puzzle difficulty is therefore a critical problem that online chess platforms need to solve at scale to optimally match many real players with adequate puzzles. To solve this challenge, we propose a chess knowledge-agnostic strategy to predict puzzle difficulty based solely on the moves of the players against their opponents trying to solve the puzzles. Specifically designed deep convolutional neural networks (CNN) were deployed as supervised learning predictors, fed with player moves represented as multichannel chessboard images. Extensive testing of our model with almost 4 million training examples against Glicko-2 evaluated puzzle difficulty ratings—considered as ground truth—resulted in good predictive performance. This was acknowledged by our runner-up result as the 7th place in the IEEE Big Data 2024 Cup and highlighted the capability of fast puzzle difficulty prediction based only on players’ moves as evidence, with no prior chess knowledge nor utilization of computationally expensive chess engines.

One characteristic feature of the traveling thief problem (TTP) is the existence of different facets of instance difficulty that make no single optimization algorithm to excel over the others. Multi-objective variants of the TTP add to the challenging features of the single-objective TTP the difficulties associated to keep a set of diverse non-dominated solutions. In this paper we propose an adaptive hybrid TTP optimization algorithm based on the probabilistic application of variation operators. The algorithm combines variation operators conceived for different facets of instance difficulty and adapts their frequency of application to the characteristics of the instances. The introduced algorithm shows a robust behavior across different bi-objective TTP instances producing results competitive with those achieved by state of the art algorithms.

Coronavirus disease 2019 (COVID-19) was first identified in respiratory samples and was found to commonly cause cough and pneumonia. However, non-respiratory symptoms including gastrointestinal disorders are also present and a big proportion of patients test positive for the virus in stools for a prolonged period. In this cross-sectional study, we investigated viral load trends in stools and nasopharyngeal swabs and their correlation with multiple demographic and clinical factors. The study included 211 laboratory-confirmed cases suffering from a mild form of the disease and completing their isolation period at a non-hospital center in the United Arab Emirates. Demographic and clinical information was collected by standardized questionnaire and from the medical records of the patient. Of the 211 participants, 25% tested negative in both sample types at the time of this study and 53% of the remaining patients had detectable viral RNA in their stools. A positive fecal viral test was associated with male gender, diarrhea as a symptom, and hospitalization during infection. A positive correlation was also observed between a delayed onset of symptoms and a positive stool test. Viral load in stools positively correlated with, being overweight, exercising, taking antibiotics in the last 3 months and blood type O. The viral load in nasopharyngeal swabs, on the other hand, was higher for blood type A, and rhesus positive (Rh factor). Regression analysis showed no correlation between the viral loads measured in stool and nasopharyngeal samples in any given patient. The results of this work highlight the factors associated with a higher viral count in each sample. It also shows the importance of stool sample analysis for the follow-up and diagnosis of recovering COVID-19 patients.

SummaryAs now well established, the world population is aging rapidly and, according to World Health Organization (WHO), the amount of people aged 60 years and older is expected to total 2 billion in 2050. For this reason, an emerging important issue is the definition of a new generation of healthcare platforms capable of monitoring people's quality of life. In this article, we propose a new methodology that supports the entire requirement elicitation process starting from the initial phase of gathering the requirements, both clinical, technological, and end‐user, up to the choice of the most suitable solution. Our proposal provides a new new iterative model in the smart healthcare field research area. Furthermore we apply our proposal in a real scenario and we report the end‐to‐end implementation of the proposed methodology.

The traveling thief problem (TTP) has emerged as a realistic multi-component problem that poses a number of challenges to traditional optimizers. In this paper we propose different ways to incorporate dynamic programming (DP) as a local optimization operator of population-based approaches to the biobjective TTP. The DP operators use different characterizations of the TTP instance to search for packing plans that improve the best current solutions. We evaluate the efficiency of the DP-based operators using TTP instances of up to 33810 cities and 338100 items, and compare the results of the DP operators with state-of-the-art algorithms for these instances. Our results show that DP-based approaches, applied individually and in combination with other types of operators, can produce good approximations of the Pareto sets for these problems.

The rapid expansion of Internet of Things (IoT) quickly evolves towards a connected network of AI-enabled, smart, multi-sensory devices that generate, consume and exchange enormous amounts of data. Stimulated by the omnipresent cloud-services and ever widening data bandwidth they aspire to control every aspect of our lives from transportation to our health, significantly increasing our reliance on the web-based services and their security. Already millions of cyber security alerts are generated every day and trigger increasingly costly investigations of the Security Operations Centres (SOC). In order to make their operation more efficient security warnings need to be reliably detected and classified based on various levels of severity, scale of potential damage or an ability to defend. We have responded to this challenge in the context of IEEE BigData Cup 2019 focused on predicting cyber security threats that require attention based on detailed logs of network activity leading to the security alert. We have developed a hybrid supervised learning ensemble model combining several state-of-the-art Extreme Gradient Boosting algorithms. Specifically, xGBoost and LightGBM model versions have been built on separate sets of features extracted from the raw logs of network events preceding the alerts and then synergically aggregated. Models' diversity imposed by algorithmic differences, complementary feature subsets, and individually optimized hyperparameters, combined with robust stratified cross-validation scheme resulted with the best true alerts detection rate yielding the AUC score in excess of 0.93, that outperformed all other 248 competing teams.

Big Data technology has discarded traditional data modeling approaches as no longer applicable to distributed data processing. It is, however, largely recognized that Big Data impose novel challenges in data and infrastructure management. Indeed, multiple components and procedures must be coordinated to ensure a high level of data quality and accessibility for the application layers, e.g., data analytics and reporting. In this paper, the third of its kind co-authored by members of IFIP WG 2.6 on Data Semantics, we propose a review of the literature addressing these topics and discuss relevant challenges for future research. Based on our literature review, we argue that methods, principles, and perspectives developed by the Data Semantics community can significantly contribute to address Big Data challenges.

Key Derivation is an important part of numerous security standards, the importance of using it was discussed throughout the literature and industry standards. On the other hand, web service security is an area that has not seen substantial research and application for Key Derivation techniques. After studying the Key Derivation techniques which are applied in Web Service Security, we find the applied algorithms and current implementations to be very limited in regard to performance and their work-flow. These limitations introduce performance bottlenecks that can limit their applicability to low power machines and mobile systems or lead to designers compromising on security to meet the quality of service desired. Moreover, this issue becomes more relevant when applied to a high performance and demanding systems such as real-time business process monitoring and messaging systems. This paper explores how Key Derivation is implemented in web services and WS-Security engines, their limitations, the performance overhead it produces and proposes an enhanced Key Derivation work-flow that takes into consideration both security and performance and allows for fine tuning them. The performance of the proposal is tested using a series of benchmarks and the security properties are verified using a well-known validation tool.