An Approach for Searching and Browsing a Network of Legal Documents

International audienceAcknowledging the importance of intertextuality in legal source analysis and the network structure of legal document collections, this paper proposes a new semantic information retrieval approach that allows for the mining of legal documents on intertextual and relational criteria. This approach is based on Formal Concept Analysis (FCA) extended with Relational Concept Analysis (RCA). It models the collection as a lattice of document classes, in which documents are clustered on the basis of their types as well as the semantic descriptors and the relations they share. This document structure allows for answering relational queries with graphs of interconnected documents and document classes, for navigating through the document lattice and for browsing the semantic and relational context of the retrieved documents. The whole approach is illustrated on a use case related to French laws and local administration acts