Abstract

Abstract
Knowledge models often use hierarchical structures, which help break down complex data into manageable components. This enables better understanding and aids in reasoning and decision-making. Hierarchical structures are effective in organizing, managing, and processing complex information. Traditional Self-Organizing Maps are typically flat, two-dimensional grids for visualizing and grouping data. They can be shaped into hierarchical structures, offering benefits such as improved data representation, scalability, enhanced grouping and visualization, and hierarchical feature extraction while preserving data topology. This paper introduces a self-organizing hierarchical map with an appropriate topology and a suitable learning mechanism for retaining information in an organized way. In this conceptual model, information is selectively absorbed in each layer. These characteristics make the Hierarchical Self-organising Maps a powerful non-linear classifier. Simulations are conducted to test and evaluate the performance of this neural structure as a classifier.

Abstract
Diabetes management data is composed of diverse factors and glycaemia indicators. Glycaemia predictive models tend to focus solely on glycaemia values. A comprehensive understanding of diabetes management requires the consideration of several aspects of diabetes management, beyond glycaemia. However, the inclusion of every aspect of diabetes management can create an overly high-dimensional data set. Excessive feature spaces increase computational complexity and may introduce over-fitting. Additionally, the inclusion of inconsequential features introduces noise that hinders a model's performance. Feature importance is a process that evaluates a feature's value, and can be used to identify optimal feature sub-sets. Depending on the context, multiple methods can be used. The drop feature method, in the literature, is considered to be the best approach to evaluate individual feature importance. To reach an optimal set, the best approach is branch and bound, albeit its heavy computational cost. This overhead can be addressed through a trade-off between the feature set's optimisation level and the process' computational feasibility. The improvement of the feature space has implications on the effectiveness of data balancing approaches. Whilst, in this study, the observed impact was not substantial, it warrants the need to reconsider the balancing approach given a superior feature space.

Abstract
This paper presents BiSAID, a dataset for exploring bipolar semantic adjectives in non-speech auditory cues, including earcons and auditory icons, i.e., sounds used to signify specific events or relay information in auditory interfaces from recorded or synthetic sources, respectively. In total, our dataset includes 599 non-speech auditory cues with different semantic labels, covering temperature (cold vs. warm), brightness (bright vs. dark), sharpness (sharp vs. dull), shape (curved vs. flat), and accuracy (correct vs. incorrect). Furthermore, we advance a preliminary analysis of brightness and accuracy earcon pairs from the BiSAID dataset to infer idiosyncratic sonic structures of each semantic earcon label from 66 instantaneous low- and mid-level descriptors, covering temporal, spectral, rhythmic, and tonal descriptors. Ultimately, we aim to unveil the relationship between sonic parameters behind earcon design, thus systematizing their structural foundations and shedding light on the metaphorical semantic nature of their description. This exploration revealed that spectral characteristics (e.g. spectral flux and spectral complexity) serve as the most relevant acoustic correlates in differentiating earcons on the dimensions of brightness and accuracy, respectively. The methodology holds great promise for systematizing earcon design and generating hypotheses for in-depth perceptual studies. © 2024. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 Unported License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original.

Abstract

Abstract
Serious games can play a crucial role in developing competencies for the job market, offering an innovative and engaging approach to learning. This study uses FLIGBY to develop employability skills among computer engineering graduates. FLIGBY is a serious game that simulates a virtual company environment where players make strategic decisions, emphasizing the principles of flow and positive psychology. Immersion in realistic scenarios provides practical experience, contributing to the development of market-ready skills. A quantitative methodology based on descriptive statistics and hypothesis testing was adopted to measure the development of competencies for the job market using the Systems Engineering Competency Framework. The results show that competence development occurs mainly in the professional and managerial dimensions. In contrast, there is no development of core and technical competencies. Furthermore, the perception of the development of these competencies occurs mainly for students with more years of professional experience. The experiential nature of FLIGBY allows users to develop practical knowledge, promoting adaptability and resilience. This gamified approach accelerates the learning curve, preparing individuals for real-world challenges in the computer engineering workplace.