Abstract
There is a growing trend to split problems into separate subproblems and develop separate models for each (e.g., different churn models for separate customer segments; different failure prediction models for separate university courses, etc.). While it may lead to better predictive models, the use of multiple models makes interpretability more challenging. In this paper, we address the problem of synthesizing the knowledge contained in a set of models without a significant loss of prediction performance. We focus on decision tree models because their interpretability makes them suitable for problems involving knowledge extraction. We detail the process, identifying alternative methods to address the different phases involved. An extensive set of experiments is carried out on the problem of predicting the failure of students in courses at the University of Porto. We assess the effect of using different methods for the operations of the methodology, both in terms of the knowledge extracted as well as the accuracy of the combined models.

Abstract

Abstract
The body's self-repair capacity is limited, including injuries on articular cartilage zones. Over the past few decades, tissue engineering and regenerative medicine (TERM) has focused its studies on the development of natural biomaterials for clinical applications aiming to overcome this self-therapeutic bottleneck. This review focuses on the development of these biomaterials using compounds and materials from marine sources that are able to be produced in a sustainable way, as an alternative to mammal sources (e.g., collagens) and benefiting from their biological properties, such as biocompatibility, low antigenicity, biodegradability, among others. The structure and composition of the new biomaterials require mimicking the native extracellular matrix (ECM) of articular cartilage tissue. To design an ideal temporary tissue-scaffold, it needs to provide a suitable environment for cell growth (cell attachment, proliferation, and differentiation), towards the regeneration of the damaged tissues. Overall, the purpose of this review is to summarize various marine sources to be used in the development of different tissue-scaffolds with the capability to sustain cells envisaging cartilage tissue engineering, analysing the systems displaying more promising performance, while pointing out current limitations and steps to be given in the near future.

Abstract
Enterprise Architecture (EA) is defined as a coherent set of principles, methods, and models used to design an organizational structure, containing business processes, information systems (IS), IT infrastructure, and other artefacts aiming the alignment of business, IT, and other organizational dimensions with the strategic objectives of a company. One of the most critical in Enterprise Architecture Management (EAM) is creating EA models representing different viewpoints for managing various company concerns on its IT landscape. At the same time, the speed of changes pressures EAM to automate modeling activities. In this context, architects need adequate tools to discover the current state of EA, enabling analyzing improvement opportunities and support architectural decisions making in a fast and agile way with more precision about the real conditions. EA Mining is the use of data mining techniques to automate the creation or update of EA models with data collected from different data sources. This work presents an exploratory review of the literature to gather the state of art on EA mining models from applications logs pursuing to automate the architecture modeling. Through this literature review, we identified the main aspects, techniques, and challenges of EA modeling automation.

Abstract

Abstract
Fluids of light is an emergent topic in optical sciences that exploits the fluid-like properties of light to establish controllable and experimentally accessible physical analogues of quantum fluids. In this work we explore this concept to generate and probe quantum turbulence phenomena by using the fluid behavior of light propagating in a defocusing nonlinear media. The proposal presented makes use of orthogonal polarizations and incoherent beam interaction to establish a theoretical framework of an analogue two-component quantum fluid, a physical system that features a modified Bogoliubov-like dispersion relation for the perturbative excitations featuring regions of instability. We demonstrate that these unstable regions can be tuned by manipulating the relative angle of incidence between the two components, allowing to define an effective range of energy injection capable of exciting turbulent phenomena. Our numerical investigations confirm the theory and show evidence of direct and inverse turbulent cascades expected from weak wave turbulence theories. The works end on a discussion concerning its possible experimental realization, allowing the access to quantum turbulence in regimes beyond those previously explored by making use of the controllable aspects of tabletop fluids of light experiments.