Dimensionality reduction is a preprocessing approach that serves to assist in reducing the number of variables (features) in the dataset, based on statistical approaches. This is extremely beneficial in machine learning (ML) applications, as it can serve to simplify the modeling process and improve the performance. In this paper, the authors apply 14 different techniques to reduce the dimensionality of a dataset with over 1200 features and test the influence across five techniques. Results indicate that some techniques have a significantly higher sensitivity to the dimensionality reduction technique applied. For example, a neural network-based technique shows a nearly 20% improvement, compared to the insignificant improvement shown by the support vector-based technique. A sensitivity difference is shown in the type of dimensionality reduction applied as well, with NB demonstrating a significant increase in performance with feature selection techniques, as opposed to decomposition techniques. Execution times are also tested, demonstrating low computational cost of most applied dimensionality reduction techniques. This indicates the need for more detailed research, to test which techniques are affected the most across different datasets.