Over the past few years, the scale of sensor networks has greatly expanded. This generates extended spatiotemporal datasets, which form a crucial information resource in numerous fields, ranging from sports and healthcare to environmental science and surveillance. Unfortunately, these datasets often contain missing values due to systematic or inadvertent sensor misoperation. This incompleteness hampers the subsequent data analysis, yet addressing these missing observations forms a challenging problem. This is especially the case when both the temporal correlation of timestamps within a single sensor and the spatial correlation between sensors are important. Here, we apply and evaluate 12 imputation methods to complete the missing values in a dataset originating from large-scale environmental monitoring. As part of a large citizen science project, IoT-based microclimate sensors were deployed for six months in 4400 gardens across the region of Flanders, generating 15-min recordings of temperature and soil moisture. Methods based on spatial recovery as well as time-based imputation were evaluated, including Spline Interpolation, MissForest, MICE, MCMC, M-RNN, BRITS, and others. The performance of these imputation methods was evaluated for different proportions of missing data (ranging from 10% to 50%), as well as a realistic missing value scenario. Techniques leveraging the spatial features of the data tend to outperform the time-based methods, with matrix completion techniques providing the best performance. Our results therefore provide a tool to maximize the benefit from costly, large-scale environmental monitoring efforts.

As global fertilizer application rates increase, high-quality datasets are paramount for comprehensive analyses to support informed decision-making and policy formulation in crucial areas such as food security or climate change. This study aims to fill existing data gaps by employing two machine learning models, eXtreme Gradient Boosting and HistGradientBoosting algorithms to produce precise country-level predictions of nitrogen (N), phosphorus pentoxide (P₂O₅), and potassium oxide (K₂O) application rates. Subsequently, we created a comprehensive dataset of 5-arcmin resolution maps depicting the application rates of each fertilizer for 13 major crop groups from 1961 to 2019. The predictions were validated by both comparing with existing databases and by assessing the drivers of fertilizer application rates using the model’s SHapley Additive exPlanations. This extensive dataset is poised to be a valuable resource for assessing fertilization trends, identifying the socioeconomic, agricultural, and environmental drivers of fertilizer application rates, and serving as an input for various applications, including environmental modeling, causal analysis, fertilizer price predictions, and forecasting.

Within the broader context of improving interactions between artificial intelligence and humans, the question has arisen regarding whether auditory and rhythmic support could increase attention for visual stimuli that do not stand out clearly from an information stream. To this end, we designed an experiment inspired by pip-and-pop but more appropriate for eliciting attention and P3a-event-related potentials (ERPs). In this study, the aim was to distinguish between targets and distractors based on the subject’s electroencephalography (EEG) data. We achieved this objective by employing different machine learning (ML) methods for both individual-subject (IS) and cross-subject (CS) models. Finally, we investigated which EEG channels and time points were used by the model to make its predictions using saliency maps. We were able to successfully perform the aforementioned classification task for both the IS and CS scenarios, reaching classification accuracies up to 76%. In accordance with the literature, the model primarily used the parietal–occipital electrodes between 200 ms and 300 ms after the stimulus to make its prediction. The findings from this research contribute to the development of more effective P300-based brain–computer interfaces. Furthermore, they validate the EEG data collected in our experiment.

In recent years, the rapid increase in available data, accompanied by significant algorithmic advancements, have enhanced our ability to analyze and extract valuable insights from this data. In this thesis, we present novel applications of machine learning (ML) in biological sciences, more particular in neuroscience, ecology, and agricultural research. We pay special attention to the interpretability and explainability of our models. ​ ​ In Chapter 2, we address a binary classification task aimed at detecting attention in electroencephalography (EEG) data. By employing state-of-the-art ML models, we successfully differentiate between target and distractor stimuli using EEG data collected during an audiovisual attention task. Additionally, we examine subject dependence in EEG data by comparing the performance of ML models trained on individual subjects versus multiple subjects. Finally, we apply explainable AI (xAI) techniques to identify the features utilized by the models for their predictions, and find that these features align with the expectations of domain experts. Chapter 3 focuses on the relationship between soil temperature, various meteorological variables, and vegetation phenology characteristics. Using ML and xAI, we find that rising soil temperatures result in an earlier onset of the growing season for plants. Additionally, our analysis reveals that the meteorological variables have the most significant impact on the vegetation phenology characteristics, while annual variations are primarily driven by changes in the soil temperature. ​ In Chapter 4, we address the issue of missing values in sensor data, with a focus on large-scale wireless sensor networks. We evaluate twelve missing value imputations methods, each using different imputation strategies and originating from diverse backgrounds. To enhance the evaluation process, we define a “masked missings” scenario, offering a more realistic assessment compared to the standard practice of using random missings. Our findings indicate that imputation methods explicitly accounting for spatial correlations between sensors generally perform best, and we use these insights to suggest directions for future research. Finally, in Chapter 5, we develop a new global dataset considering the historical application of fertilizers. Specifically, we use ML to predict historical fertilizer application at both the crop and country levels, based on a set of features related to crop classes, along with socioeconomic, agrological, and environmental variables. Additionally, we use xAI to identify the most relevant drivers influencing fertilizer application.

Changes in climate can greatly affect the phenology of plants, which can have important feedback effects, such as altering the carbon cycle. These phenological feedback effects are often induced by a shift in the start or end dates of the growing season of plants. The normalized difference vegetation index (NDVI) serves as a straightforward indicator for assessing the presence of green vegetation and can also provide an estimation of the plants' growing season. In this study, we investigated the effect of soil temperature on the timing of the start of the season (SOS), timing of the peak of the season (POS), and the maximum annual NDVI value (PEAK) in subarctic grassland ecosystems between 2014 and 2019. We also explored the impact of other meteorological variables, including air temperature, precipitation, and irradiance, on the inter-annual variation in vegetation phenology. Using machine learning (ML) techniques and SHapley Additive exPlanations (SHAP) values, we analyzed the relative importance and contribution of each variable to the phenological predictions. Our results reveal a significant relationship between soil temperature and SOS and POS, indicating that higher soil temperatures lead to an earlier start and peak of the growing season. However, the Peak NDVI values showed just a slight increase with higher soil temperatures. The analysis of other meteorological variables demonstrated their impacts on the inter-annual variation of the vegetation phenology. Ultimately, this study contributes to our knowledge of the relationships between soil temperature, meteorological variables, and vegetation phenology, providing valuable insights for predicting vegetation phenology characteristics and managing subarctic grasslands in the face of climate change. Additionally, this work provides a solid foundation for future ML-based vegetation phenology studies.