Japanese Information Extraction with Automatically Extracted Patterns

One of the central issues for informa- tion extraction (IE) systems is the cost of customization from one scenario to another. Research on the automated acquisition of patterns is important for portability and scalability. This pa- per explores the automatic extraction of patterns in Japanese from unannotated text. We introduce two modules of our system, the pattern extraction module and the information extraction module, both of which use structural patterns. The performance of the whole system is measured in a MUC-style evaluation.