String Vector based KNN for Index Optimization

In this research, we propose the string vector based KNN as the approach to the index optimization. The task may be viewed into an instance of word classification and the problems in encoding words or texts into numerical vectors were solved by encoding texts into string vectors in the previous works on text mining tasks. Influence by the previous works, we encode words into string vectors, as well as texts, define the semantic operations on string vectors, and apply the modified version of the KNN to the index optimization which is mapped into a classification task. As the benefits from this research, we may expect the better performance and more compact representations than encoding texts or words into numerical vectors. Therefore, the goal of this research is to develop the index optimization system with the benefits.