K Nearest Neghbor modified into String Vector based Version for Keyword Extraction

This article proposes the modified KNN (K Nearest Neighbor)algorithm which receives a string vector as its input data and isapplied to the keyword extraction. The results from applying thestring vector based algorithms to the text categorizations weresuccessful in previous works and the keyword extraction is able tobe mapped into the binary classification where each word isclassified into keyword or non-keyword. In the proposed system, atext which is given as the input is indexed into a list of words,each word is classified by the proposed KNN version, and the wordswhich are classified into keyword are extracted ad the output. Theproposed KNN version is empirically validated as the better approachin deciding whether each word is a keyword or non-keyword in newsarticles and opinions. We need to define and characterizemathematically more operations on string vectors for modifying moreadvanced machine learning algorithms.