Table based KNN for index optimization

Abstract

We concern this research with the table based KNN as the approach to the index optimization task. It may be interpreted into an instance of word classification, and the encoding scheme where words are encoded into tables improved the task word classification. In this research, words are encoded into tables and apply the table based KNN to the index optimization task. From this research, we expect the better and more stable performance than the traditional version, in this task. Therefore, this research is intended to provide the improved index optimization scheme.