Categorical Attribute traNsformation Environment (CANE) is a simpler but powerful data categorical preprocessing Python package. The package is valuable since there is currently a large range of Machine Learning (ML) algorithms that can only be trained using numerical data (e.g., Deep Learning, Support Vector Machines) and several real-world ML applications are associated with categorical data attributes. Currently, CANE offers three categorical to numeric transformation methods, namely: Percentage Categorical Pruned (PCP), Inverse Document Frequency (IDF) and a simpler One-Hot-Encoding method. Additionally, the CANE module is well documented with several code examples that can help in its adoption by non expert users.

This research work focuses on machine learning algorithms in order to make predictions in financial markets. The foremost objective is to test whether the two machine learning algorithms: SVM and LSTM are capable of predicting the price movement in different time-frames and then develop a comparison analysis. In this research work, it is applied supervised machine learning with different input features. The practical and software component of this thesis applies Python programming language to test the hypothesis and act as proof of concept. The financial data quotes were obtained through online financial databases. The results demonstrate that SVM is capable of predicting the direction of the price while the LSTM did not present reliable results.

The city of Lisbon has complex geology and is subject to significant natural hazards of geological and climatic origin, such as earthquakes, floods, and mass movements, which requires a deeper understanding of the subsurface in the context of urban planning. However, the available geological information is mostly two-dimensional and sporadic, which limits its applicability. Three-dimensional modelling allows for the integration of multiple data sources, reducing uncertainties and supporting more informed and sustainable decisions. This approach contributes to a holistic understanding of urban geology and the definition of resilience strategies. Currently, there is no city-scale 3D geological model at the national level. The work done is for a small area in Lisbon, using closed-source computer applications, which makes it difficult to access or replicate. In addition, the available information on the data and methodology to be used is scattered, so this report sought to conduct exhaustive research on the subject. Research was also conducted on existing geological-geotechnical models in Portugal and worldwide, highlighting their impact on society. The aim of this work is to design and build a three-dimensional model of a pilot area located in the city of Lisbon, with a view to providing a geometric representation of the geological formations. In this work, the pilot area to be considered for the design of the model, the geological formations present in it, and the available information were defined. Three different methodologies were tested: surveys, profiles, and outcrop patterns, using ArcGIS Pro and Python programming. Based on the results presented, it was concluded that the best method would be the outcrop pattern, using the thicknesses of the formations and the slope. The three-dimensional geological model then materialized in voxel format and constitutes the geometric basis for entering subsurface parameterization data.