Automated Concept Extraction to aid Legal eDiscovery Review
Prasad Deshpande, Thomas Hampp, Manjula Hosurmath, Sachindra Joshi, Seema Meena · 2009
E-Discovery is the process of discovering electronically stored information such as email that is relevant to a legal case. A typical e-discovery process incurs huge costs due to the large volume of information and the requirement of highly specialized and expensive human resources (legal professionals). In this paper, we examine how information management technologies can be used to reduce the high cost. We propose a set of concepts that are helpful in identifying relevant and not-relevant documents. We then develop a set of rule based annotators that automatically identify documents with these concepts and compare their performance with standard off-the-shelf classifiers for building the concept annotators. The rule based annotators have been integrated into the IBM product for e-discovery review called IBM InfoSphere eDiscovery Analyzer. 1.