In the context of preparing a dataset for a machine learning project, you are tasked with exploring the relationships between multiple numerical variables to identify potential correlations and outliers. Considering the need for efficiency and clarity in visualization, which of the following visualization techniques would be most appropriate for this task? Choose the best option. (A) To display the distribution of a single variable, (B) To visualize hierarchical data, (C) To show trends over time, (D) To visualize the relationship between multiple pairs of variables