This article proposes the modified KNN (K Nearest Neighbor) algorithm which considers the feature similarity and is applied to the keyword extraction. The texts which are given as features for encoding words into numerical vectors are semantic related entities, rather than independent ones, and the keyword extraction is able to be viewed into a binary classification where each word is classified into keyword or non-keyword. In the proposed system, a text which is given as the input is indexed into a list of words, each word isclassified by the proposed KNN version, and the words which are classified into keyword are extracted ad the output. The proposed KNN version is empirically validated as the better approach in deciding whether each word is a keyword or non-keyword in news articles and opinions. The significance of this research is to improve the classification performance by utilizing the feature similarities.