Feature selection for model calibration

Lucia Gomez Teijeiro

In today’s information era, data analysis often involves dealing with vast amounts of information, commonly known as Big Data. One way to streamline this process is through the use of Feature Selection (FS) methods. These techniques enable the reduction of features in a dataset while retaining sufficient information to achieve better modeling performance. In this context, we will examine the benefits and potential drawbacks associated with FS, taking into consideration the analytical objective and the intrinsic characteristics of the data. Although numerous FS techniques exist, we will only discuss a select few, such as those utilizing dimensionality reduction or graphs, that can serve as valuable tools in enhancing data analysis.