Feature engineering

<p>Feature engineering is a preprocessing step in <a href="/facts/Supervised_machine_learning/JlNFNPFt">supervised machine learning</a> and <a href="/facts/Statistical_modeling/jKkT5ftm">statistical modeling</a> which transforms raw data into a more effective set of inputs. Each input comprises several attributes, known as features. By providing models with relevant information, feature engineering significantly enhances their predictive accuracy and decision-making capability. 
</p><p>Beyond machine learning, the principles of feature engineering are applied in various scientific fields, including physics. For example, physicists construct <a href="/facts/Dimensionless_quantity/TINrpQIq">dimensionless numbers</a> such as the <a href="/facts/Reynolds_number/4sdungsf">Reynolds number</a> in <a href="/facts/Fluid_dynamics/y7uzSCfg">fluid dynamics</a>, the <a href="/facts/Nusselt_number/WHGa4D7r">Nusselt number</a> in <a href="/facts/Heat_transfer/1IV3WbFL">heat transfer</a>, and the <a href="/facts/Archimedes_number/gVLxQqoT">Archimedes number</a> in <a href="/facts/Sedimentation/qUgHGvhd">sedimentation</a>. They also develop first approximations of solutions, such as analytical solutions for the <a href="/facts/Strength_of_materials/CE77cUcB">strength of materials</a> in mechanics.
</p>

Feature engineering open-in-new

Feature engineering