Abstract
Arbitrary, inconsistent, or faulty decision-making raises serious concerns, and preventing unfair models is an increasingly important challenge in Machine Learning. Data often reflect past discriminatory behavior, and models trained on such data may reflect bias on sensitive attributes, such as gender, race, or age. One approach to developing fair models is to preprocess the training data to remove the underlying biases while preserving the relevant information, for example, by correcting biased labels. While multiple label noise correction methods are available, the information about their behavior in identifying discrimination is very limited. In this work, we develop an empirical methodology to systematically evaluate the effectiveness of label noise correction techniques in ensuring the fairness of models trained on biased datasets. Our methodology involves manipulating the amount of label noise and can be used with fairness benchmarks but also with standard ML datasets. We apply the methodology to analyze six label noise correction methods according to several fairness metrics on standard OpenML datasets. Our results suggest that the Hybrid Label Noise Correction [20] method achieves the best trade-off between predictive performance and fairness. Clustering-Based Correction [14] can reduce discrimination the most, however, at the cost of lower predictive performance.

Abstract

Abstract
Towards the decarbonisation of the power system, digital substations have gradually increased in smart grids, where Ethernet cables have replaced large quantities of copper wires. With this transition, the standardised communication protocols through the LAN network play a central role in exchanging information and data between the physical power system and the control centres. One of the well-known protocols in the digital substations is IEC 61850 GOOSE (Generic Object-Oriented Substation Event), which is used to share time-critical information related to protection, automation, and control. The transmission time of this protocol affects power system operation and raises various issues, such as communication latencies and incorrect information. Therefore, it is necessary to consider the protocol transmission time for further protection and control mechanisms to ensure the stability and efficiency of the power system. For this purpose, this paper contributes the implementation of a cyber–physical power system (CPPS) testbed to measure the transfer time of IEC 61850 GOOSE under the real-time domain using the real-time simulator, Typhoon HIL, and its toolchains. This paper can benefit scholars and researchers in the relevant domains in implementing a CPPS testbed and an approach for transfer time measurement of communication protocols within the laboratory, eliminating the need for real-world substation devices.

Abstract
This paper investigates the real-time tracking capabilities of the Long Baseline (LBL) acoustic tracking system, developed by INESC TEC, for near-shore monitoring applications with human divers. The study aims to determine the system's suitability for environmental monitoring under challenging conditions, such as very shallow water and areas close to the coastline, where acoustic multipath effects are prevalent and can cause significant measurement errors. To mitigate these errors, a two-stage outlier rejection algorithm was implemented. The algorithm's performance was evaluated by comparing the measurement data at each stage and assessing the reduction in erroneous readings. The tracking performance was evaluated based on accuracy and repeatability. Two dives were performed, during which positions marked using the developed system were compared with GNSS data. © 2024 IEEE.

Abstract
This study adopts an integrative review approach to explore the differences and similarities between smart cities and sustainable cities. The research starts by performing two systematic literature reviews about both paradigms and, after that, employs a thematic analysis to identify key themes, definitions, and characteristics that differentiate and connect these two urban development concepts. The findings reveal more similarities than differences between the two paradigms. Despite this, some key differences are identified. Smart cities are characterized by their use of advanced information and communication technologies to enhance urban infrastructure, improve public services, and optimize resource management. In contrast, sustainable cities focus on environmental conservation, social equity, and economic viability to ensure long-term urban resilience and quality of life. This study is important because it clarifies both concepts and highlights the potential for integrating smart and sustainable city strategies to address contemporary urban challenges more holistically. The findings also suggest a convergence towards the concept of 'smart sustainable cities', which leverage technology to achieve sustainability goals. Finally, this study concludes by identifying research gaps and proposing a future research agenda to further understand and optimize the synergy between smart and sustainable urban development paradigms.

Abstract
Diabetic foot is a complication that carries a considerable risk in diabetic patients. The consequent loss of protective sensitivity in the lower limbs requires an early diagnosis due to the imminent possibility of ulceration or amputation of the affected limb. To assess the loss of protective sensitivity, the 10 gf Semmes-Weinstein (SW) monofilament is the most used first-line procedure. However, the used device is most often non-calibrated and its feedback can lead to decision errors. In this paper we present an equipment that is able to automatically conduct a metrological verification and evaluation of the 10 gf SW monofilament in the assessment of the loss of protective sensitivity. Additionally, the proposed equipment is able to simulate the practicioner's procedure, or can be used for training purposes, providing force-feedback information. After calibration, displacement vs. buckling force contours were plotted for three distinct monofilaments, confirming then ability of the equipment to provide fast, detailed and precise information.