Tech moves fast! Stay ahead of the curve with Techopedia!
Join nearly 200,000 subscribers who receive actionable tech insights from Techopedia.
In machine learning, feature selection is the use of specific variables or data points to maximize efficiency in this type of advanced data science.
Feature selection is also known as variable selection, attribute selection or subset selection.
With feature selection, engineers and data scientists are able to tune out a lot of the “noise” in a given system. Using feature selection helps to discard redundant or irrelevant data, and this culling can make the machine learning results stronger. For instance, in a marine biology project, researchers could use feature selection to only select certain classifying information on one or more surveyed species, and eliminate other data that is not central to the project.
Feature selection can be done with various types of tools including Weka, Scikit-learn and R. This can help create more accurate models, and generally improve machine learning processes. Engineers have to work with feature selection and training data to prevent overfitting and other problems. Feature selection also helps teams to avoid the “curse of dimensionality,” which is a shorthand for certain kinds of data problems in complex computing operations.