A Native Topic Map Database Management System and Its N-gram Index for Efficient Search of Topics

A topic map has composed a topic expressing concepts, an association between the concepts and a role of association. A topic map database is currently existed to be utilized to a system. The existing Topic map database (TMDB) uses a system such as relational database and object oriented database, and it was necessary to convert from the structure of each system to a topic map structure. Therefore, we propose a native TMDB management system that can store and manipulate data without changing the data structure. Topics connected directly in adjacency list using memory mapped files realize to search at high speed. In addition, we propose a method to speed up the search of the topic based on its nature.