<p>A number of factors impact the comfort of our lives. Among these are meteorological factors, such as relative humidity, which plays a pivotal role in determining our state of comfort. Despite the vast number of papers published annually on this topic, the scientific community has yet to fully elucidate the consequences of relative humidity on our health and our environment. This manuscript presents an overview of the effects of this parameter on our health. Following a full consideration of the impacts of relative humidity, the objective of the study is to predict hygrothermal comfort days based on relative humidity and temperature. Days with high diseases risk are classified as 0, while days with optimal comfort are classified as 1. Three classification methods were tested: logistic regression, deep neural networks, and the Sugeno Adaptive Neuro-Fuzzy Inference System. The performance of the models is evaluated by calculating and comparing the classification metrics (accuracy, precision, F1 score and recall). Three groups of predictors were employed in the study: one comprising temperature and relative humidity as predictors, another incorporating six variables (solar irradiance, direct solar irradiance, precipitation, evapotranspiration, vapor pressure deficit, and wind speed), and a third including vapor pressure deficit and relative humidity. The findings demonstrate that deep neural network (DNN) models, achieving superior to 98% accuracy with temperature and relative humidity in comparison to logistic regression, alternative deep learning models, and the Sugeno-based ANFIS. This superiority is evident from high decision metric values, well-generalized decision boundaries, and optimal confusion matrix distributions.</p>