Recently, the way information is accessed has changed significantly because of the rapid growth of the Internet and a move toward electronic publishing. With that condition, there will be a need to efficiently determine and extract relevant information from digital documents. This study has proposed that template mining can be used for extracting different kinds of information from digital documents, and described research of reference linking system and then built template for citation information using analysis result of sample articles. Also, this study has shown the potential of template mining to automatically create citation databases.