Automatic generation of the keywords dictionary for efficient document form identification

摘要

On the task of Document Form Identification by using keywords matching, it is essential to choose appropriate set of keywords for the accurate identification of each document. So far, well trained operators selected appropriate keywords by try and error. But as the number of document increases, it becomes tremendously laborious task. In this paper, we propose the method to choose keywords automatically by using the measure of stability and specificity of each keyword. By the method, we could save the cost of choosing keywords without degrading the rate of correct identification of documents.