Domain N-Gram Construction and Its Application to Text Editor

Google has published n-gram data which was constructed from huge document set gathered until 2005. However, it is hard to use the data in real world applications due to its huge volume. In this paper, we propose a method to construct domain n-gram data in which a specific domain group is interested and apply the data to text editor for practical efficiency in evaluation. It contains diverse test results according to typing speed level of people and comparison results with other works. The result of this research is conducted through applying to typing only however it has big importance in a point of being capable of expecting its effectiveness because the n-gram data is widely applicable to many fields.