NIMS: MagIC - Capítulos de livros
URI permanente para esta coleção:
Navegar
Entradas recentes
- Interconnectedness in education systemsPublication . Candia, Cristian; Pulgar, Javier; Pinheiro, Flávio L.; Information Management Research Center (MagIC) - NOVA Information Management School; NOVA Information Management School (NOVA IMS)Underlying complex systems is a rich web of interconnected components that determine the relational properties of the system. Yet, traditional methods used in education sciences often disregard the underlying complexity of the educational system and, consequently, its emergence phenomena. Here, we argue that an interconnected vision of educational systems—from classrooms to an organizational level—is key to improving learning, social integration, well-being, and decision-making, all fundamental aspects of the educational experience. Understanding the education system as an interconnected network of people, degree programs, and institutions requires methods and concepts from computational social science. Thus, we can leverage institutional records and (quasi) experimental designs to elicit the relational maps of key players in education and derive their implications in their functioning at all scales. In different settings, from elementary classrooms to higher education programs, we show how mapping the network relationships between entities can lead to the inference of novel insights about education systems and the development of solutions with societal implications.
- Exploring Non-bloating Geometric Semantic Genetic ProgrammingPublication . Vanneschi, Leonardo; Farinati, Davide; Rasteiro, Diogo; Rosenfeld, Liah; Pietropolli, Gloria; Silva, Sara; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management SchoolRecently, a new variant of Geometric Semantic Genetic Programming(GSGP) was introduced that, while maintaining the property of inducing a unimodal error surface for all supervised learning problems, is able to generate models that are compact enough to be interpretable by humans. This variant is called the Semantic Learning algorithm based on Inflate and deflate Mutation (SLIM_GSGP) and, as the name suggests, it is based on two types of mutation: one (inflate) that generates offspring larger than their parents, similar to traditional geometric semantic mutation, and the other (deflate) that generates offspring smaller than their parents. This chapter reviews and extends the initial work on SLIM_GSGP by introducing two novel variants, thoroughly studying the geometric characteristics of the SLIM_GSGP operators and discussing the experimental results and their interpretation in greater depth. The main conclusion is that SLIM_GSGP is a very promising method, warranting significant investment in future research.
- Contribution Towards Smart CitiesPublication . Bação, Fernando José Ferreira Lucas; Henriques, Roberto; Antunes, Jorge; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management SchoolThe interest in using information to improve the quality of living in large urban areas and the efficiency of its governance has been around for decades. Nevertheless, recent developments in information and communications technology have sparked new ideas in academic research, all of which are usually grouped under the umbrella term of Smart Cities. The concept of Smart City can be defined as cities that are lived, managed and developed in an information-saturated environment. However, there are still several significant challenges that need to be tackled before we can realize this vision. In this study we aim at providing a small contribution in this direction, by maximizing the usefulness of the already available information resources. One of the most detailed and geographically relevant information resources available for studying cities is the census, more specifically, the data available at block level. In this study we use self-organizing maps (SOM) to explore the block level data included in the 2001 and 2011 Portuguese censuses for the city of Lisbon. We focus on measuring change, proposing new ways to compare the two time periods, which have two different underlying geographical bases. We proceed with the analysis of the data using different SOM variants, aiming at providing a twofold portrait: showing how Lisbon evolved during the first decade of the twenty-first century and how both the census dataset and the SOMs can be used to produce an informational framework for micro analysis of urban contexts.
- Spatial Autocorrelation and Association MeasuresPublication . Negreiros, J.; Painho, Marco; Lopes, I.; Costa, A. C.; NOVA Information Management School (NOVA IMS)Several classical statements relating to the definition of GIS can be found in specialized literature such as the GIS International Journal, expressing the idea that spatial analysis can somehow be useful. GIS is successful not only because it integrates data, but it also enables us to share data in different departments or segments of our organizations. I like this notion of putting the world’s pieces back together again (ArcNews, 2000). “GIS is simultaneously the telescope, the microscope, the computer and the Xerox machine of regional analysis and the synthesis of spatial data” (Abler, 1988). “GIS is a system of hardware, software and liveware implemented with the aim of storing, processing, visualizing and analyzing data of a spatial nature. Other definitions are also possible” (Painho, 1999). “GIS is a tool for revealing what is otherwise invisible in geographical information” (Longley, Goodchild, Maguire, & Rhind, 2001). Certainly, GIS is not a graphic database.
- Participatory Geographic Information SystemsPublication . de Sá, Dulce Magalhães; Costa, Ana Cristina M.; NOVA Information Management School (NOVA IMS)Geographic information systems are largely used in different levels of administration and planning where geo-referenced information is a crucial factor behind analysis and determination of different decision-making scenarios. The potential of these systems is increasingly being perceived as a support to facilitate public participation in planning processes.
- Economic Complexity and Inequality at the National and Regional LevelsPublication . Hartmann, Dominik; Pinheiro, Flávio L.; Information Management Research Center (MagIC) - NOVA Information Management School; NOVA Information Management School (NOVA IMS)Recent studies have found evidence of a negative association between economic complexity and inequality at the country level. Moreover, evidence suggests that sophisticated economies tend to outsource products that are less desirable (e.g. in terms of wage and inequality effects), and instead focus on complex products requiring networks of skilled labor and more inclusive institutions. Yet the negative association between economic complexity and inequality on a coarse scale could hide important dynamics at a fine-grained level. Complex economic activities are difficult to develop and tend to concentrate spatially, leading to "winner-take-most" effects that spur regional inequality in countries. Large, complex cities tend to attract both high- and low-skill activities and workers, and are also associated with higher levels of hierarchies, competition, and skill premiums. As a result, the association between complexity and inequality reverses at regional scales; in other words, more complex regions tend to be more unequal. Ideas from polarization theories, institutional changes, and urban scaling literature can help to understand this paradox, while new methods from economic complexity and relatedness can help identify inclusive growth constraints and opportunities.
- Large Language Models (LLMs) for Smart Manufacturing and Industry X.0Publication . Baptista, Marcia L.; Yue, Nan; Islam, M. M. Manjurul; Prendinger, Helmut; Information Management Research Center (MagIC) - NOVA Information Management School; NOVA Information Management School (NOVA IMS)The manufacturing industry is rapidly changing, creating a growing demand for more intelligent and adaptive systems. With recent developments in artificial intelligence, especially with the onset of large language models (LLMs) such as ChatGPT, new opportunities have emerged for companies to increase their productivity and maximize revenue. In a competitive environment, businesses must constantly innovate to stay ahead. To support innovative and competitive organizations, LLMs can analyze large amounts of data to identify trends and optimize processes. In addition, the industry faces a labor shortage, particularly in roles that require specialized skills. LLMs can fill this gap by providing real-time assistance and training. This knowledge transfer could help less experienced workers perform their tasks more effectively. Regulatory compliance is increasingly imperative in manufacturing, and LLMs can help ensure adherence to safety standards and regulatory requirements. LLMs can address these and other challenges by using their capabilities in data processing, natural language understanding, and predictive analytics. In this chapter, we explain the fundamental concepts behind LLM techniques and how to use them in a smart manufacturing environment such as Industry X.0. We discuss the challenges and future trends of LLMs in different industrial fields. We also highlight the need for LLM frameworks that can guarantee data privacy, security, and ethical usage.
- MARFRONTPublication . Goulding, Samuel; Painho, Marco; Gil, Fernando; Information Management Research Center (MagIC) - NOVA Information Management School; NOVA Information Management School (NOVA IMS)Evaluating the Portuguese coast to determine its permeability to irregular entry via the sea Immigration is and always has been a factor that shapes politics, economics, and cultures at an international level. In Europe, the number of irregular entries along the continent’s external border reached a record high in 2014, registering around 280,000 migrants crossing the EU border illegally. As an EU member, Portugal has been very active in not only participating in joint operations, but also in the development of new methodologies for monitoring and detecting irregular migrants, mainly through the testing of drones from a Portuguese airbase in the southern part of the country (the Algarve). At the national level, Portugal does not have a problem with immigration, but this article seeks to evaluate the Portuguese coast to determine its permeability to irregular entry via the sea. This evaluation was developed as a possible future tool that Portuguese border guards could use to identify the most critical areas along the Portuguese shoreline. Though it is not the case right now, Portugal could become the next country of entry for migrants leaving North Africa due to the ever-increasing levels of monitoring, detection, and overall security on the Mediterranean Sea. Portugal’s coastline is 943 kilometers long, but for this analysis, we are only taking into account the south and southwest coastline (the Algarve and Alentejo regions), assuming an individual trying to enter Portuguese territory from the North of Africa would try a disembark in this area.
- From conceptualisation to actionPublication . Poli, Susi; Oliveira, Cristina; Zsár, Virág; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management SchoolThis chapter examines various definitions and perceptions of Research Management and Administration (RMA) from individuals both from within and outside the profession to gain a wider understanding of this field. These definitions and perceptions are expected to trigger reflections on where the boundaries of the profession are more likely to be. To do so, the authors utilise a mixed method that begins with a discussion of different definitions of RMA. Next, we move from conceptualisation to action and engage the reader by presenting empirical insights from an analysis of specific training programmes within RMA, shedding light on the profession's distinctive features from an insider's perspective. Lastly, we delve into the case study of the project foRMAtion, a training program that introduces RMAs as the 'Professionals at the Interface of Science.' This case study allows us to explore how individuals outside the RMA profession, such as teachers and students participating in its training courses, perceive and understand RMA.
- Credit Risk ScoringPublication . Raimundo, Bernardo; Bravo, Jorge M.; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management SchoolForecasting the creditworthiness of customers in new and existing loan contracts is a central issue of lenders’ activity. Credit scoring involves the use of analytical methods to transform historical loan application and loan performance data into credit scores that signal creditworthiness, inform, and determine credit decisions, determine credit limits, and loan rates, and assist in fraud detection, delinquency intervention, or loss mitigation. The standard approach to credit scoring is to pursue a “winner-take-all” perspective by which, for each dataset, a single believed to be the “best” statistical learning or machine learning classifier is selected from a set of candidate approaches using some method or criteria often neglecting model uncertainty. This paper empirically investigates the predictive accuracy of single-based classifiers against the stacking generalization approach in credit risk modelling using real-world peer-to-peer lending data. The findings show that stacking ensembles consistently outperform most traditional individual credit scoring models in predicting the default probability. Moreover, the findings show that adopting a feature selection process and hyperparameter tuning contributes to improving the performance of individual credit risk models and the super-learner scoring algorithm, helping models to be simpler, more comprehensive, and with lower classification error rates. Improving credit scoring models to better identify loan delinquency can substantially contribute to reducing loan impairments and losses leading to an improvement in the financial performance of credit institutions.