Nearest centroid classifier

<p>In <a href="/facts/Machine_learning/e0w0XJTu">machine learning</a>, a nearest centroid classifier or nearest prototype classifier is a <a href="/facts/Statistical_classification/jXXHRkXR">classification model</a> that assigns to observations the label of the class of training samples whose <a href="/facts/Mean/swcEd4Pg">mean</a> (<a href="/facts/Centroid/H6dDbg0h">centroid</a>) is closest to the observation. When applied to <a href="/facts/Text_classification/B69lL6BH">text classification</a> using <a href="/facts/Vector_space_model/WC1LqLA5">word vectors</a> containing <a href="/facts/Tf*idf/JDzc5Jnr">tf*idf</a> weights to represent documents, the nearest centroid classifier is known as the Rocchio classifier because of its similarity to the <a href="/facts/Rocchio_algorithm/a5fWdBxm">Rocchio algorithm</a> for <a href="/facts/Relevance_feedback/bDYwe063">relevance feedback</a>.
</p><p>An extended version of the nearest centroid classifier has found applications in the medical domain, specifically classification of <a href="/facts/Tumor/K0ytzmPz">tumors</a>.
</p>

Nearest centroid classifier open-in-new

Nearest centroid classifier