This thesis presents an applied research for the needs of a company Datlowe, s.r.o. aimed at automatic processing of legal documents. The goal of the work is to design, implement and evaluate a classification module that is able to assign categories to the paragraphs of the documents. Several classification algorithms are used, evaluated and compared to each other to be consequently combined to obtain the best models. The outcome is a prediction module which was successfully integrated into the entire document processing system. Other contributions, along with the classification module, are the measurement of the inter-annotator agreement and introducing new set of features for classification.

Legal knowledge engineering methodology for epistemologically sound, large scale legal expert systems is developed in this dissertation. A specific meta-epistemological method is posed for the transformation of legal domain epistemology to large scale legal expert systems; the method has five stages: 1. domain epistemology; 2. computational domain epistemology; 3. shell epistemology; 4. programming epistemology; and 5. application epistemology and ontology. The nature of legal epistemology is defined in terms of a deep model that divides the information of the ontology of legal possibilities into the three sorts of logic premises, namely, (1) rules of law for extended deduction, (2) material facts of cases for induction that establishes rule antecedents, and (3) reasons for rules, including justifications, explanations or criticisms of rules, for abduction. Extended deduction is distinguished for automation, and provides a map for locating, relatively, associated induction and abduction. Added to this is a communication system that involves issues of cognition and justice in the legal system. The Appendix sets out a sample of draft rule maps of the United Nations Convention on Contracts for the International Sale of Goods, known as the Vienna Convention, to illustrate that the substantive epistemology of the international law can be mapped to the generic epistemology of the shell. This thesis deflects the ontological solution back to the earlier rule-based, case-based and logic advances, with a definition of artificial legal intelligence that rests on legal epistemology; added to the definition is a transparent communication system of a user interface, including an interactive visualisation of rule maps, and the heuristics that process input and produce output to give effect to the legal intelligence of an application. The additions include an epistemological use of the ontology of legal possibilities to complete legal logic, for the purposes of processing specific legal applications. While the specific meta-epistemological methodology distinguishes domain epistemology from the epistemologies of artificial legal intelligence, namely computational domain epistemology, program design epistemology, programming epistemology and application epistemology, the prototypes illustrate the use of those distinctions, and the synthesis effected by that use. The thesis develops the Jurisprudence of Legal Knowledge Engineering by an artificial metaphysics. / Doctor of Philosophy (Ph.D)

[pt] Como uma consequência natural da digitalização do sistema judiciário brasileiro, um grande e crescente número de documentos jurídicos tornou-se disponível na internet, especialmente decisões judiciais. Como ilustração, em 2020, o Judiciário brasileiro produziu 25 milhões de decisões. Neste mesmo ano, o Supremo Tribunal Federal (STF), a mais alta corte do judiciário brasileiro, produziu 99.5 mil decisões. Alinhados a esses valores, observamos uma demanda crescente por estudos voltados para a extração e exploração do conhecimento jurídico de grandes acervos de documentos legais. Porém, ao contrário do conteúdo de textos comuns (como por exemplo, livro, notícias e postagem de blog), o texto jurídico constitui um caso particular de uso de uma linguagem altamente convencionalizada. Infelizmente, pouca atenção é dada à extração de informações em domínios especializados, como textos legais. Do ponto de vista temporal, o Judiciário é uma instituição em constante evolução, que se molda para atender às demandas da sociedade. Com isso, o nosso objetivo é propor um processo confiável de extração de informações jurídicas de grandes acervos de documentos jurídicos, tomando como base o STF e as decisões monocráticas publicadas por este tribunal nos anos entre 2000 e 2018. Para tanto, pretendemos explorar a combinação de diferentes técnicas de Processamento de Linguagem Natural (PLN) e Extração de Informação (EI) no contexto jurídico. Da PLN, pretendemos explorar as estratégias automatizadas de reconhecimento de entidades nomeadas no domínio legal. Do ponto da EI, pretendemos explorar a modelagem dinâmica de tópicos utilizando a decomposição tensorial como ferramenta para investigar mudanças no raciocinio juridico presente nas decisões ao lonfo do tempo, a partir da evolução do textos e da presença de entidades nomeadas legais. Para avaliar a confiabilidade, exploramos a interpretabilidade do método empregado, e recursos visuais para facilitar a interpretação por parte de um especialista de domínio. Como resultado final, a proposta de um processo confiável e de baixo custo para subsidiar novos estudos no domínio jurídico e, também, propostas de novas estratégias de extração de informações em grandes acervos de documentos. / [en] As a natural consequence of the Brazilian Judicial System’s digitization, a large and increasing number of legal documents have become available on the Internet, especially judicial decisions. As an illustration, in 2020, 25 million decisions were produced by the Brazilian Judiciary. Meanwhile, the Brazilian Supreme Court (STF), the highest judicial body in Brazil, alone has produced 99.5 thousand decisions. In line with those numbers, we face a growing demand for studies focused on extracting and exploring the legal knowledge hidden in those large collections of legal documents. However, unlike typical textual content (e.g., book, news, and blog post), the legal text constitutes a particular case of highly conventionalized language. Little attention is paid to information extraction in specialized domains such as legal texts. From a temporal perspective, the Judiciary itself is a constantly evolving institution, which molds itself to cope with the demands of society. Therefore, our goal is to propose a reliable process for legal information extraction from large collections of legal documents, based on the STF scenario and the monocratic decisions published by it between 2000 and 2018. To do so, we intend to explore the combination of different Natural Language Processing (NLP) and Information Extraction (IE) techniques on legal domain. From NLP, we explore automated named entity recognition strategies in the legal domain. From IE, we explore dynamic topic modeling with tensor decomposition as a tool to investigate the legal reasoning changes embedded in those decisions over time through textual evolution and the presence of the legal named entities. For reliability, we explore the interpretability of the methods employed. Also, we add visual resources to facilitate interpretation by a domain specialist. As a final result, we expect to propose a reliable and cost-effective process to support further studies in the legal domain and, also, to propose new strategies for information extraction on a large collection of documents.