Legal content fusion for legal information retrieval

With recent increasing attention to legal information processing, legal information retrieval (IR) has become one of the active research fields. However, there are still many hindrances obtaining rigorous results in legal IR applications in comparison with IR applications for general document retrieval. It is mainly due to the characteristics of legal information such as the complicated structure of legal contents and usage of legal jargon. In this paper, we present a legal IR method, which is a structure-wise IR approach. e presented method in this study focuses on analyzing the contents of legal documents and applying the content contributions to the IR processing. We demonstrate the performance of the proposed IR method with the COILEE data set, which are derived from Japanese bar exams.