An overview of memory: some issues on structures and organization in the legal domain

Lawyers often need to look for previous similar legal cases when analysing new ones. The more previous cases, the more time is spent. Classical search engines execute termbased retrieval, which may miss relevant documents as well as fetch several irrelevant ones, causing lack of useful information and waste of time. Ideally, retrieval should be meaning-based. Humans beings are able to do e cient searches due to their knowledge. Therefore, semantic search requires knowledge. This paper presents a semantic search engine. Along the paper, several issues concerning specially knowledge representation and memory are discussed. A formalism based on models of comprehension is introduced, as well as its motivation. Examples of representation of sentences in natural language from the Legal Domain are provided. The search engine and its architecture, based on domain knowledge, are brie y commented. The main goal is to give legal o ces the opportunity to save time by providing a more suitable document retrieval. Eje: Agentes y Sistemas Inteligentes (ASI) Red de Universidades con Carreras en Informática (RedUNCI)