Knowledge-Based Legal Document Retrieval: A Case Study on Italian Civil Court Decisions

In this paper, we present a knowledge-based approach for legal document retrieval based on the organization of a textual data repository and on document embedding models. Pre-processed and embedded documents are iteratively classified at sentence level through a terminology extraction and concept formation cycle, using a zero-knowledge approach that offers a high degree of flexibility with regard to the integration of external knowledge and the variability of inputs, suitable to face the scarcity of annotated data and the specificity of terminology that feature the Italian legal domain document corpora.