Hier-SPCNet: A Legal Statute Hierarchy-based Heterogeneous Network for Computing Legal Case Document Similarity

Computing similarity between two legal case documents is an important and\nchallenging task in Legal IR, for which text-based and network-based measures\nhave been proposed in literature. All prior network-based similarity methods\nconsidered a precedent citation network among case documents only (PCNet).\nHowever, this approach misses an important source of legal knowledge -- the\nhierarchy of legal statutes that are applicable in a given legal jurisdiction\n(e.g., country). We propose to augment the PCNet with the hierarchy of legal\nstatutes, to form a heterogeneous network Hier-SPCNet, having citation links\nbetween case documents and statutes, as well as citation and hierarchy links\namong the statutes. Experiments over a set of Indian Supreme Court case\ndocuments show that our proposed heterogeneous network enables significantly\nbetter document similarity estimation, as compared to existing approaches using\nPCNet. We also show that the proposed network-based method can complement\ntext-based measures for better estimation of legal document similarity.\n