Abstract
By leveraging the advances in wireless communications networks and their ubiquitous nature, sensing through communication technologies has flourished in recent years. In particular, Human-to-Machine Interfaces have been exploiting WiFi IEEE 802.11 networks to obtain information that allows Human Activity Recognition. In this paper, we propose a classification model to perform Person Identification (PI) through Body Velocity Profile time series, obtained by combining Channel State Information containing gesture knowledge from multiple Access Points. Through this model, we investigate the impact of different gestures on PI classification performance and explore how informing the model about the input gesture can enhance classification accuracy. This information may enable the network to adjust to the absence of features capable of adequately characterizing the desired classes in certain gestures. A simplified stacking model is also presented, capable of combining the softmax outputs of K previously proposed individual models. By having the individual models’ evaluations of a gesture and the gesture information relating to it, the number of gestures considered was shown to significantly improve the performance of the PI classification task. This enhancement increased 17% of the average F1 scores when compared to the individual model tested on the same data.

Abstract
Recent device-level cyber-attacks have targeted IoT critical applications in power distribution systems integrated with the Internet communications infrastructure. These systems utilize group domain of interpretation (GDOI) as designated by International Electrotechnical Commission (IEC) power utility standards IEC 61850 and IEC 62351. However, GDOI cannot protect against novel threats, such as IoT device-level attacks that can modify device firmware and configuration files to create command and control malicious communication. As a consequence, the attacks can compromise substations with potentially catastrophic consequences. With this in mind, this article proposes a permissioned/private blockchain-based authentication framework that provides a solution to current security threats such as the IoT device-level attacks. Our work improves the GDOI protocol applied in critical IoT applications by achieving decentralized and distributed device authentication. The security of our proposal is demonstrated against known attacks as well as through formal mechanisms via the joint use of the AVISPA and SPAN tools. The proposed approach adds negligible authentication latency, thus ensuring appropriate scalability as the number of nodes increases. Our work addresses the problem of device-level cyber-attacks such as device identity theft and introduction of fake nodes in GDOI-enabled smart grids. It introduces a permissioned blockchain based device authentication management in the GDOI phase 1 and periodic device integrity check in phase 2 to achieve decentralized authentication and device-level security. image

Abstract
With the proliferation of Internet of Things (IoT) devices, there is an increasing need to prioritize their security, especially in the context of identity and authentication mechanisms. However, IoT devices have unique limitations in terms of computational capabilities and susceptibility to hardware attacks, which pose significant challenges to establishing strong identity and authentication systems. Paradoxically, the very hardware constraints responsible for these challenges can also offer potential solutions. By incorporating hardware-based identity implementations, it is possible to overcome computational and energy limitations, while bolstering resistance against both hardware and software attacks. This research addresses these challenges by investigating the vulnerabilities and obstacles faced by identity and authentication systems in the IoT context, while also exploring potential technologies to address these issues. Each identified technology underwent meticulous investigation, considering known security attacks, implemented countermeasures, and an assessment of their pros and cons. Furthermore, an extensive literature survey was conducted to identify instances where these technologies have effectively supported device identity. The research also includes a demonstration that evaluates the effectiveness of hardware trust anchors in mitigating various attacks on IoT identity. This empirical evaluation provides valuable insights into the challenges developers encounter when implementing hardware-based identity solutions. Moreover, it underscores the substantial value of these solutions in terms of mitigating attacks and developing robust identity frameworks. By thoroughly examining vulnerabilities, exploring technologies, and conducting empirical evaluations, this research contributes to understanding and promoting the adoption of hardware-based identity and authentication systems in secure IoT environments. The findings emphasize the challenges faced by developers and highlight the significance of hardware trust anchors in enhancing security and facilitating effective identity solutions.

Abstract
Learning how to program is a difficult task. To acquire the re-quired skills, novice programmers must solve a broad range of programming activities, always supported with timely, rich, and accurate feedback. Automated assessment tools play a major role in fulfilling these needs, being a common pres-ence in introductory programming courses. As programming exercises are not easy to produce and those loaded into these tools must adhere to specific format requirements, teachers often opt for reusing them for several years. There-fore, most automated assessment tools, particularly Mooshak, store hundreds of submissions to the same programming ex-ercises, as these need to be kept after automatically pro-cessed for possible subsequent manual revision. Our dataset consists of the submissions to 16 programming exercises in Mooshak proposed in multiple years within the 2003-2020 timespan to undergraduate Computer Science students at the Faculty of Sciences from the University of Porto. In particular, we extract their code property graphs and store them as CSV files. The analysis of this data can enable, for instance, the generation of more concise and personalized feedback based on similar accepted submissions in the past, the identifica-tion of different strategies to solve a problem, the under -standing of a student's thinking process, among many other findings.(c) 2023 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ )

Abstract
Learning to program requires diligent practice and creates room for discovery, trial and error, debugging, and concept mapping. Learners must walk this long road themselves, supported by appropriate and timely feedback. Providing such feedback in programming exercises is not a humanly feasible task. Therefore, the early and steadily growing interest of computer science educators in the automated assessment of programming exercises is not surprising. The automated assessment of programming assignments has been an active area of research for over a century, and interest in it continues to grow as it adapts to new developments in computer science and the resulting changes in educational requirements. It is therefore of paramount importance to understand the work that has been performed, who has performed it, its evolution over time, the relationships between publications, its hot topics, and open problems, among others. This paper presents a bibliometric study of the field, with a particular focus on the issue of automatic feedback generation, using literature data from the Web of Science Core Collection. It includes a descriptive analysis using various bibliometric measures and data visualizations on authors, affiliations, citations, and topics. In addition, we performed a complementary analysis focusing only on the subset of publications on the specific topic of automatic feedback generation. The results are highlighted and discussed.

Abstract
Class imbalance is a common issue while developing classification models. In order to tackle this problem, synthetic data have recently been developed to enhance the minority class. These artificially generated samples aim to bolster the representation of the minority class. However, evaluating the suitability of such generated data is crucial to ensure their alignment with the original data distribution. Utility measures come into play here to quantify how similar the distribution of the generated data is to the original one. For tabular data, there are various evaluation methods that assess different characteristics of the generated data. In this study, we collected utility measures and categorized them based on the type of analysis they performed. We then applied these measures to synthetic data generated from two well-known datasets, Adults Income, and Liar+. We also used five well-known generative models, Borderline SMOTE, DataSynthesizer, CTGAN, CopulaGAN, and REaLTabFormer, to generate the synthetic data and evaluated its quality using the utility measures. The measurements have proven to be informative, indicating that if one synthetic dataset is superior to another in terms of utility measures, it will be more effective as an augmentation for the minority class when performing classification tasks.