Data, Knowledge, and Discovery in Legal Cases
David D. Lewis Consulting
Abstract: The past decade has seen the birth of a multi-billion dollar electronic discovery (e-discovery) software and services industry in which information retrieval (IR) and artificial intelligence (AI) methods play a pivotal role. Machine learning and statistical IR play a central role, and I will review their applications in e-discovery, as well as the perceptions that have emerged about them in the e-discovery community. I then take up two very different ways in which the demands of e-discovery suggest new research areas for AI. First, I make the case that knowledge-based methods should, and are likely to, play a larger role in e-discovery in the future. Second, I outline a new computational learning theory framework (Finite Population Annotation) motivated by the nature of machine learning applications in e-discovery, and the need for their outputs to stand up in court.
Biography: David D. Lewis, Ph.D. (www.DavidDLewis.com) is a Chicago-based consulting computer scientist working in the areas of information retrieval, data mining, natural language processing, and the evaluation of complex information systems. He formerly held research positions at AT&T Labs, Bell Labs, and the University of Chicago. He has published more than 75 scientific papers and 8 patents, and was elected a Fellow of the American Association for the Advancement of Science in 2006.
Rules-Driven Business Services: Flexibility within the Boundaries of the Law
Jorke van der Pol
Senior advisor, Ministry of the Interior and Kingdom Relations Immigration and Naturalisation Service, The Netherlands
Abstract: The Dutch Immigration and Naturalization Service (IND) is renewing its organization, business processes and complete IT landscape in the Renewal Program “IND with the Times”. One aspect of this program is the new information system INDiGO. This is a knowledge-based, service-oriented system, composed of standard components (Siebel, FileNet, Oracle Fusion Middleware) and a rule engine (BeInformed). The rule engine contains all knowledge on relevant laws, regulations and policies and offers this knowledge ‘as-a-service’. But the rule engine also contains knowledge on the process flow, i.e. the order in which the 165 Business Services are to be executed. At any moment in time, the rule-engine can be consulted to provide those Business Services that are relevant to the specific case and customer at that specific moment. This creates a highly flexible system; changes in law, regulation or procedures can be changed and implemented very quickly through the use of this rule engine. Moreover, the system is event driven; events that have an impact on the flow (the decision making process) can be evaluated as they occur. Last but not least, INDiGO is designed to support the highly skilled, professional end-user, who remains in control at all times. In this presentation the focus will be on the guiding principles behind the information system (the separation of the “know” from the “flow”), the approach and methodology employed, the implementation strategy and, of course, the lessons learned.
Biography: Jorke van der Pol is a senior advisor at the Dutch Immigration and Naturalization Services. Currently he is involved in solving various business challenges with an IT component. Making full use of his expertise and on-the-ground experience of immigration and naturalisation law and procedures, he searches for (technological) solutions in the fields of rule governance. His approach can be characterized as agile and result-oriented. He is involved in various projects within the INDiGO project.
Models of the Law in AI & Law
Faculty of Law of the University of Bologna, Cirsfid, and European University Institute of Florence
Abstract: I will address the different ways in which the law has been and may be seen by the AI & law community. First I will consider what aspects of the law are most relevant to the main lines of inquiry of AI & law, such as information retrieval, rule-based reasoning, case-based reasoning, ontologies, argumentation, theory construction and multi-agent systems. I will argue that a rich picture of the law is emerging from the AI & law research, which in some respects goes beyond the accounts currently provided by legal theory. Then, I will address norm-based societies of autonomous agents, where agents themselves have to identify what norms apply to them, in what context, and may be able to produce the norms governing their interaction. I will argue that to model normative reasoning in such contexts we need to address the dynamical nature of normative systems, namely, the way in which social facts change the content of normative systems, and how agents can refer to such facts for determining what norms they are subject to. Finally, I shall consider how agents, when faced with the request of multiple normative systems, can reason using multiple normative references and taking into account intra-systemic connections.
Biography: Giovanni Sartor is professor of Legal informatics and Legal Theory at the Faculty of Law of the University of Bologna - Cirsfid, and at European University Institute of Florence. He obtained a PhD at the European University Institute (Florence), worked at the Court of Justice of the European Union (Luxembourg), was a researcher at the Italian National Council of Research (ITTIG, Florence), held the chair in Jurisprudence at Queen’s University of Belfast. He is current President of the International Association for Artificial Intelligence and Law. He has published widely on legal reasoning, argumentation, computational logic, legislation technique, and computer law.
Beyond Jeopardy!: Prospects for DeepQA in Legal Text Analysis
Language Technologies Institute, School of Computer Science, Carnegie Mellon University
Abstract: The success of IBM's Watson system in the Jeopardy! challenge has raised the public's awareness of question answering technology, and the prospects for its application in real-world domains such as health care and law. This presentation combines an overview of the DeepQA software architecture that powers Watson with a discussion of possible areas of application for the legal profession.
Biography: Dr. Eric Nyberg is a Professor in the Language Technologies Institute in the School of Computer Science at Carnegie Mellon University. Since joining the CMU faculty in 1993, his research has focused on practical applications of language technologies in areas such as machine translation and question answering. Most recently, Dr. Nyberg collaborated with IBM's David Ferrucci and others to found the Open Advancement of Question Answering initiative, a framework for industry-university collaboration, which seeks to advance the use of question answering technology in multiple application domains.