A major challenge of today’s society is to make large urban centres more sustainable. Improving the energy efficiency of the various infrastructures that make up cities is one aspect being considered when improving their sustainability, with Wastewater Treatment Plants (WWTPs) being one of them. Consequently, this study aims to conceive, tune, and evaluate a set of candidate deep learning models with the goal being to forecast the energy consumption of a WWTP, following a recursive multi-step approach. Three distinct types of models were experimented, in particular, Long Short-Term Memory networks (LSTMs), Gated Recurrent Units (GRUs), and uni-dimensional Convolutional Neural Networks (CNNs). Uni- and multi-variate settings were evaluated, as well as different methods for handling outliers. Promising forecasting results were obtained by CNN-based models, being this difference statistically significant when compared to LSTMs and GRUs, with the best model presenting an approximate overall error of 630 kWh when on a multi-variate setting. Finally, to overcome the problem of data scarcity in WWTPs, transfer learning processes were implemented, with promising results being achieved when using a pre-trained uni-variate CNN model, with the overall error reducing to 325 kWh.

Traffic flow forecasting is an acknowledged time series problem whose solutions have been essentially grounded on statistical-based models. Recent times came, however, with promising results regarding the use of Recurrent Neural Networks (RNNs), such as Long Short-Term Memory networks (LSTMs), to accurately address time series problems. Literature is, however, evasive in regard to several aspects of the conceived models and often exhibits misconceptions that may lead to important pitfalls. This study aims to conceive and find the best possible LSTM model for traffic flow forecasting while addressing several important aspects of such models such as the multitude of input features, the time frames used by the model and the employed approach for multi-step forecasting. To overcome the spatial problem of open source datasets, this study presents and describes a new dataset collected by the authors of this work. After several weeks of model fitting, Recursive Multi-Step Multi-Variate models were the ones showing better performance, strengthening the perception that LSTMs can be used to accurately forecast the traffic flow for several future timesteps.

Goldberg’s 100 Unipolar Markers remains one of the most popular ways to measure personality traits, in particular, the Big Five. An important reduction was later preformed by Saucier, using a sub-set of 40 markers. Both assessments are performed by presenting a set of markers, or adjectives, to the subject, requesting him to quantify each marker using a 9-point rating scale. Consequently, the goal of this study is to conduct experiments and propose a shorter alternative where the subject is only required to identify which adjectives describe him the most. Hence, a web platform was developed for data collection, requesting subjects to rate each adjective and select those describing him the most. Based on a Gradient Boosting approach, two distinct Machine Learning architectures were conceived, tuned and evaluated. The first makes use of regressors to provide an exact score of the Big Five while the second uses classifiers to provide a binned output. As input, both receive the one-hot encoded selection of adjectives. Both architectures performed well. The first is able to quantify the Big Five with an approximate error of 5 units of measure, while the second shows a micro-averaged f1-score of 83%. Since all adjectives are used to compute all traits, models are able to harness inter-trait relationships, being possible to further reduce the set of adjectives by removing those that have smaller importance.