Legal text retrieval traditionally relies upon external knowledge sources such as thesauri and classification schemes, and an accurate indexing of the documents is often manually done. As a result not all legal documents can be effectively retrieved. However a number of current artificial intelligence techniques are promising for legal text retrieval. They sustain the acquisition of knowledge and the knowledge-rich processing of the content of document texts and information need, and of their matching. Currently, techniques for learning information needs, learning (...) concept attributes of texts, information extraction, text classification and clustering, and text summarization need to be studied in legal text retrieval because of their potential for improving retrieval and decreasing the cost of manual indexing. The resulting query and text representations are semantically much richer than a set of key terms. Their use allows for more refined retrieval models in which some reasoning can be applied. This paper gives an overview of the state of the art of these innovativetechniques and their potential for legal text retrieval. (shrink)

This paper examines the use of connectionism (neural networks) in modelling legal reasoning. I discuss how the implementations of neural networks have failed to account for legal theoretical perspectives on adjudication. I criticise the use of neural networks in law, not because connectionism is inherently unsuitable in law, but rather because it has been done so poorly to date. The paper reviews a number of legal theories which provide a grounding for the use of neural networks in law. It then (...) examines some implementations undertaken in law and criticises their legal theoretical naïvete. It then presents a lessons from the implementations which researchers must bear in mind if they wish to build neural networks which are justified by legal theories. (shrink)

Much legal research focuses on understanding how judicial decision-makers exercise their discretion. In this paper we examine the notion of legal or judicial discretion, and weaker and stronger forms of discretion. At all times our goal is to build cognitive models of the exercise of discretion, with a view to building computer software to model and primarily support decision-making. We observe that discretionary decision-making can best be modeled using three independent axes: bounded and unbounded, defined and undefined, and binary and (...) continuous. Examples of legal tasks are given from each of the eight ensuing octants and we conclude by saying what this model shows about current legal trends. We should stress that our taxonomy has been based on our observations of how discretionary legal decisions are made. No claim is made that our model is either complete (providing advice in every domain) or exact, but it does help knowledge engineers construct legal decision support systems in discretionary domains. (shrink)

Hypertext may represent a new paradigm capable of exploring legal sources within which links are established according to pertinent relationships found between statute texts and case law. However, to discover relevant information in such a network, a browsing mechanism is not enough when faced with a large volume of texts. This paper describes a new retrieval model where documents are represented according to both their content and relationships with other sources of information.

This paper argues the thesis that a particular style of reasoning, qualitative comparative reasoning (QCR), plays a role in at least three areas of legal reasoning that are central in AI and law research, namely legal theory construction, case-based reasoning in the form of case comparison, and legal proof. The paper gives an informal exposition of one particular way to deal with QCR, based on the author’s previous work on reason-based logic (RBL). Then it contains a substantially adapted formalisation of (...) RBL, to make RBL suitable for dealing with QCR. The paper concludes with a brief discussion of related work. (shrink)

This paper considers the problem of how to evaluate an offender’s criminal record. This evaluation is part of the sentencing process carried out by a judge, and may be complicated in the case of offenders with a heavy record. We give a comprehensive overview of the approach to an offender’s past record in various (Western) countries, considering the two major approaches: desert-based and utilitarian. The paper describes the determination of the parameters involved in the evaluation, and the construction of a (...) decision support system to be used by a judge about to pass sentence in a criminal case. The system may be used as a stand-alone system, but can also be integrated as a component of a general sentencing support system, e.g. a statistical sentencing information system. The system is a knowledge-based system. The knowledge base (in rule-form) was created primarily by eliciting knowledge from experts (judges, lawyers, academics and probation officers), but also by applying statutory law, case law and legal authoritative texts. It is flexible in the sense, that a user can introduce his1 own preferences (expressed as numerical coefficients) and thereby apply different sentencing principles. The prototype is at present undergoing testing by some of the participating judges. (shrink)

This paper aims at providing an account of juridical acts that forms a suitable starting point for the creation of computational systems that deal with juridical acts. The paper is divided into two parts. Because juridical acts will be analyzed as intentional changes in the world of law, the ‘furniture’ of this world, that consists broadly speaking of entities, facts and rules, plays a central role in the analysis. This first part of the paper deals with this furniture and its (...) philosophical underpinnings, and at the same time introduces most of the logical apparatus that will be used to deal with it. The focus in the first part is on static and dynamic legal rules and their interplay in constituting the world of law. (shrink)