Abstract
The still prevalent use of paper conformity lists in the automotive industry has a serious negative impact on the performance of quality control inspectors. We propose instead a hybrid quality inspection system, where we combine automated detection with human feedback, to increase worker performance by reducing mental and physical fatigue, and the adaptability and responsiveness of the assembly line to change. The system integrates the hierarchical automatic detection of the non-conforming vehicle parts and information visualization on a wearable device to present the results to the factory worker and obtain human confirmation. Besides designing a novel 3D vehicle generator to create a digital representation of the non conformity list and to collect automatically annotated training data, we apply and aggregate in a novel way state-of-the-art domain adaptation and pseudo labeling methods to our real application scenario, in order to bridge the gap between the labeled data generated by the vehicle generator and the real unlabeled data collected on the factory floor. This methodology allows us to obtain, without any manual annotation of the real dataset, an example-based F1 score of 0.565 in an unconstrained scenario and 0.601 in a fixed camera setup (improvements of 11 and 14.6 percentage points, respectively, over a baseline trained with purely simulated data). Feedback obtained from factory workers highlighted the usefulness of the proposed solution, and showed that a truly hybrid assembly line, where machine and human work in symbiosis, increases both efficiency and accuracy in automotive quality control.

Abstract
Research data management (RDM) practices are critical for ensuring research success. Data can assume diverse formats and data in image format have been understudied in RDM. To understand image management habits in research, we have conducted semi-structured interviews with researchers from four research domains. Most researchers do not formally manage their images, nor do they develop RDM plans. They assume that image management is not a topic discussed at project meetings. In turn, they tend to perform some individual practices, depending on the context and their own opinion, such as creating captions to describe the images and organizing and storing the images in specific locations. However, they see these habits as necessary and admit that they will start to do so in a formal and collaborative way with the working group. These results provide valuable information on practical aspects of the use and production of images in research.

Abstract
The use of mobile conversations is increasing all around the world. A conversational agent (CA) is mostly useful due to the fast response times and their simple nature. Recently, we have seen the development and increasing use of dialog systems on the Web. A conversational agent (CA) is a system capable of conversing with a user in natural language, in a way that it simulates a human dialog. Examples of CA can be found in several areas, including healthcare, entertainment, business, and education. In this paper a state of the art review of these dialog systems is presented, comprising different categories, different approaches and trends. The purpose of this work is to identify and compare the main existing approaches for building CA, categorizing them and highlighting the main strengths and weaknesses. Furthermore, it seeks to contextualize their use in an educational context and to discover the issues related to this task that may help in the choice of future investigations in the area of conversational natural language processing in educational context.

Abstract
Pressure ulcers are a critical issue not only for patients, decreasing their quality of life, but also for healthcare professionals, contributing to burnout from continuous monitoring, with a consequent increase in healthcare costs. Due to the relevance of this problem, many hardware and software approaches have been proposed to ameliorate some aspects of pressure ulcer prevention and monitoring. In this article, we focus on reviewing solutions that use sensor-based data, possibly in combination with other intrinsic or extrinsic information, processed by some form of intelligent algorithm, to provide healthcare professionals with knowledge that improves the decision-making process when dealing with a patient at risk of developing pressure ulcers. We used a systematic approach to select 21 studies that were thoroughly reviewed and summarized, considering which sensors and algorithms were used, the most relevant data features, the recommendations provided, and the results obtained after deployment. This review allowed us not only to describe the state of the art regarding the previous items, but also to identify the three main stages where intelligent algorithms can bring meaningful improvement to pressure ulcer prevention and mitigation. Finally, as a result of this review and following discussion, we drew guidelines for a general architecture of an intelligent pressure ulcer prevention system.

Abstract
O presente artigo estuda a influência dos fatores socioeconómicos dos diferentes municípios no sucesso educacional dos estudantes. Para verificar a existência de fatores relevantes para o percurso académico dos estudantes, foram obtidos datasets com descritores socioeconómicos por município, médias das notas dos exames nacionais e as taxas de sucesso dos alunos. Estes datasets foram submetidos a uma técnica de K-nearest neighbours para permitir encontrar valores de atributos em municípios com valores em falta. Foram, de seguida, aplicados algoritmos de classificação, através de árvores de decisão e regressão, que permitiram analisar quais os atributos socioeconómicos que tinham, potencialmente, maior relação com o sucesso escolar. O trabalho efetuado permite identificar alguns fatores como alvos de potenciais estudos futuros sem, no entanto, se verificar correlações fortes com nenhum atributo socioeconómico.;This paper studies the influence of the socio-economic factors of different municipalities on the educational success of students. To verify the existence of relevant factors to the academic course of the students, datasets were obtained with socio-economic descriptors by municipality, average grades of national exams and success rates of students. These datasets were submitted to a K-nearest neighbours technique to allow finding attributes in municipalities with missing values. Classification algorithms were then applied through decision and regression trees, which allowed analyzing which socio-economic attributes were potentially more related to school success. The work performed allowed identifying some factors as targets of potential future studies without, however, verifying strong correlations with any socio-economic attribute.

Abstract
Human interaction with intelligent systems, services, and devices generates large volumes of user-related data. This multi-source information can be used to build richer user profiles and improve personalization. Our goal is to combine multi-source data to create user profiles by assigning dynamic individual weights. This paper describes a multi-source user profiling methodology and illustrates its application with a film recommendation system. The contemplated data sources include (i) personal history, (ii) explicit preferences (ratings), and (iii) social activities (likes, comments, or shares). The MovieLens dataset was selected and adapted to assess our approach by comparing the standard and the proposed methodologies. In the standard approach, we calculate the best global weights to apply to the different profile sources and generate all user profiles accordingly. In the proposed approach, we determine, for each user, individual weights for the different profile sources. The approach proved to be an efficient solution to a complex problem by continuously updating the individual data source weights and improving the accuracy of the generated personalised multimedia recommendations. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.