Abstract
Since its appearance in 2008, blockchain technology has found multiple uses in fields such as banking, supply chain management, and healthcare. One of the most intriguing uses of blockchain is in voting systems, where the technology can overcome the security and transparency concerns that plague traditional voting systems. This paper provides a thorough examination of the implementation of a blockchain-based voting system. The proposed system employs cryptographic methods to protect voters' privacy and anonymity while ensuring the verifiability and integrity of election results. Digital signatures, homomorphic encryption (He), zero-knowledge proofs (ZKPs), and the Byzantine fault-tolerant consensus method underpin the system. A review of the literature on the use of blockchain technology for voting systems supports the analysis and the technical and logistical constraints connected with implementing the suggested system. The study suggests solutions to problems such as managing voter identification and authentication, ensuring accessibility for all voters, and dealing with network latency and scalability. The suggested blockchain-based voting system can provide a safe and transparent platform for casting and counting votes, ensuring election results' privacy, anonymity, and verifiability. The implementation of blockchain technology can overcome traditional voting systems' security and transparency shortcomings while also delivering a high level of integrity and traceability.

Abstract
Recommender systems position themselves as powerful tools in the support of relevance and personalization, presenting remarkable potential in the area of marketing. The cold-start customer problematic presents a challenge within this topic, leading to the need of distinguishing user features and preferences based on a restricted set of transactional information. This paper proposes a hybrid recommender system that aims to leverage transactional and portfolio information as indicating characteristics of customer behaviour. Four independent systems are combined through a parallelised weighted hybrid design. The first individual system utilises the price, target age, and brand of each product to develop a content-based recommender system, identifying item similarities. Secondly, a keyword-based content system uses product titles and descriptions to identify related groups of items. The third system utilises transactional data, defining similarity between products based on purchasing patterns, categorised as a collaborative model. The fourth system distinguishes itself from the previous approaches by leveraging association rules, using transactional information to establish antecedent and precedence relationships between items through a market basket analysis. Two datasets were analysed: product portfolio and transactional datasets. The product portfolio had 17,118 unique products and the included 4,408,825 instances from 2 June 2021 until 2 June 2022. Although the collaborative system demonstrated the best evaluation metrics when comparing all systems individually, the hybridisation of the four systems surpassed each of the individual systems in performance, with a 8.9% hit rate, 6.6% portfolio coverage, and with closer targeting of customer preferences and smaller bias.

Abstract
Identifying stress in people is not a trivial or straightforward task, as several factors are involved in detecting the presence or absence of stress. The problem of detect stress has attracted much attention in the last decade and is mainly addressed with physiological signals and in a controlled ambience with specific tasks. However, the widespread use of video cameras permitted the creation of a new non-invasive data collection techniques. The goal of this work is to provide an alternative way to detect stress in the workplace without the need of specific laboratory conditions. For that, a stress detection model based on images analysed with deep learning neural networks was developed. The trained model achieved a F1 = 79.9% on a binary dataset, of stress/non-stress, with an imbalanced ratio of 0.49. This model can be used in a non-invasive application to detect stress and provide recommendations to the collaborators in the workplace in order to help them to control their stress condition. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract
A terceira edição dos Cadernos de Investigação do Mestrado em Negócio Eletrónico (MNE) testemunha o contínuo amadurecimento deste ciclo de estudos como polo de reflexão académica e científica. Este volume reúne 21 artigos de jovens investigadores que, sob orientação de docentes-investigadores, exploram os fenómenos mais relevantes que moldam o atual panorama do negócio eletrónico.

Abstract
O Centro de Estudos Organizacionais e Sociais do Politécnico do Porto (CEOS.PP) juntou-se à iniciativa anual da Comissão Europeia - "Noite Europeia dos Investigadores" – lançada em 2005 - com o Dia do Investigador CEOS [DICEOS23]. O objetivo deste evento, que decorreu no dia 29 de setembro de 2023, foi o de divulgar o trabalho desenvolvido pelos investigadores do CEOS.PP, um momento que contou com um conjunto de atividades para criar sinergias entre os investigadores deste centro de investigação e abrir caminhos para o futuro.

Abstract
The optimal Coalition Structure Generation (CSG) problem for a given set of agents finds a partition of the agent set that maximises social welfare. The CSG problem is an NP-hard optimisation problem, where the search space grows exponentially. The exact and approximation algorithms focus on finding an optimal solution or a solution within a known bound from the optimum. However, as the number of agents increases linearly, the search space increases exponentially and a practical option here is to use heuristic algorithms. Heuristic algorithms are suitable for solving the optimisation problems because of their less computational complexity. TACOS is a heuristic method for the CSG problem that finds high-quality solutions quickly using a neighbourhood search performed with a memory. However, some of the neighbourhood searches by TACOS can be performed simultaneously. Therefore, this paper proposes a parallel version of the TACOS algorithm (P-TACOS) for the CSG problem, intending to find a better solution than TACOS. We evaluated P-TACOS using eight (8) benchmark data distributions. Results show that P-TACOS achieves better results for all eight (8) data distributions. P-TACOS achieves the highest gain, 74.23%, for the Chisquare distribution and the lowest gain, 0.01%, for the Normal distribution. We also examine how often P-TACOS generates better results than TACOS. In the best case, it generates better results for 92.30% of the time (for the Rayleigh and Agent-based Normal distributions), and in the worst case, 38.46% of the time (for the Weibull distribution).