Application of citation network analysis for improved similarity index estimation of legal case documents : A study

With the availability of information in online databases, information retrieval has become back bone of many processes today. Significant share of this online data comprises of unstructured and textual data. There are many domains like legal domain which rely solely on information stored in various legal documents. Legal domain is considered to be very complex and its processes are largely dependent on knowledge interpretation by human expert. Establishing relevance and similarity between two cases based on expositions in various legal documents is the most common but non trivial task performed by a legal expert. This paper discusses application of network analysis to compare two approaches for finding legal document similarity - a) Cosine similarity b) Citation based similarity. Results show that citation based similarity measure is more robust in determining parallel among cases.