Modification of K Nearest Neighbor by Graph Similarity Metric for Keyword Extraction

This article proposes the modified KNN (K Nearest Neighbor)algorithm which receives a graph as its input data and is applied tothe keyword extraction. The graph is more graphical for representinga word and the keyword extraction is able to be mapped into thebinary classification where each word is classified into keyword ornon-keyword. In the proposed system, a text which is given as theinput is indexed into a list of words, each word is classified bythe proposed KNN version, and the words which are classified intokeyword are extracted ad the output. The proposed KNN version isempirically validated as the better approach in deciding whethereach word is a keyword or non-keyword in news articles and opinions.In this article, a word is encoded into a weighted and undirectedgraph and it is represented into a list of edges.