Kevin D. Ashley

Reasoning with cases and hypotheticals in HYPO

Abstract HYPO is a case-based reasoning system that evaluates problems by comparing and contrasting them with cases from its Case Knowledge Base (CKB). It generates legal arguments citing the past cases as justifications for legal conclusions about who should win in problem disputes involving trade secret law. HYPOs arguments present competing adversarial views of the problem and it poses hypotheticals to alter the balance of the evaluation. HYPO uses Dimensions as a generalization scheme for accessing and evaluating cases. HYPOs reasoning process and various computational definitions are described and illustrated, including its definitions for computing relevant similarities and differences, the most on point and best cases to cite, four kinds of counter-examples, targets for hypotheticals and the aspects of a case that are salient in various argument roles. These definitions enable HYPO to make contextually sensitive assessments of relevance and salience without relying on either a strong domain theory or a priori weighting schemes.

A case-based system for trade secrets law

In this paper, we give an overview of our case-based reasoning program, HYPO, which operates in the field of trade secret law. We discuss key ingredients of case-based reasoning, in general, and the correspondence of these to elements of HYPO. We conclude with an extended example of HYPO working through a hypothetical trade secrets case, patterned after an actual case.

Textual case-based reasoning

This commentary provides a definition of textual case-based reasoning (TCBR) and surveys research contributions according to four research questions. We also describe how TCBR can be distinguished from text mining and information retrieval. We conclude with potential directions for TCBR research.

A case-based approach to modeling legal expertise

Hypo, a computer program that performs case-based reasoning in the legal domain, helps attorneys analyze and make arguments about new fact situations in terms of the most relevant precedent cases. To perform this task, Hypo must make factual comparisons of cases relative to the problem situation and determine the legal significance of comparisons in terms of arguments about the problem situation. The authors describe techniques that Hypo uses to compare cases, choose the best cases for evaluating, and construct arguments about a new fact situation. They demonstrate how Hypo critically compares a problem situation to the most relevantly similar precedent cases to outline an argument regarding how to decide the current fact situation (CFS) based on its significant similarities to and differences from most on point cases (MOPCS). Hypos main tool for this task is the claim lattice mechanism. The authors present a detailed example of a claim lattice actually generated by Hypo to analyze a real legal case.<<ETX>>

Predicting outcomes of case based legal arguments

In this paper, we introduce IBP, an algorithm that combines reasoning with an abstract domain model and case-based reasoning techniques to predict the outcome of case-based legal arguments. Unlike the predictions generated by statistical or machine-learning techniques, IBPs predictions are accompanied by explanations.We describe an empirical evaluation of IBP, in which we compare our algorithm to prediction based on Hypos and CATOs relevance criteria, and to a number of widely used machine learning algorithms. IBP reaches higher accuracy than all competitors, and hypothesis testing shows that the observed differences are statistically significant. An ablation study indicates that both sources of knowledge in IBP contribute to the accuracy of its predictions.

The Role of Information Extraction for Textual CBR

The benefits of CBR methods in domains where cases are text depend on the underlying text representation. Today, most TCBR approaches are limited to the degree that they are based on efficient, but weak IR methods. These do not allow for reasoning about the similarities between cases, which is mandatory for many CBR tasks beyond text retrieval, including adaptation or argumentation. In order to carry out more advanced CBR that compares complex cases in terms of abstract indexes, NLP methods are required to derive a better case representation. This paper discusses how state-of-the-art NLP/IE methods might be used for automatically extracting relevant factual information, preserving information captured in text structure and ascertaining negation. It also presents our ongoing research on automatically deriving abstract indexing concepts from legal case texts. We report progress toward integrating IE techniques and ML for generalizing from case texts to our CBR case representation.

Improving the representation of legal case texts with information extraction methods

The prohibitive cost of assigning indices to textual cases is a major obstacle for the practical use of AI and Law systems supporting reasoning and arguing with cases. While progress has been made toward extracting certain facts from well-structured case texts or classifying case abstracts under Key Number concepts, these methods still do not suffice for the complexity of indexing concepts in CBR systems. In this paper, we lay out how a better example representation may facilitate classification-based indexing. Our hypotheses are that (1) abstracting from the individual actors and events in cases, (2) capturing actions in multi-word features, and (3) recognizing negation, can lead to a better representation of legal case texts for automatic indexing. We discuss how to implement these techniques with state-of-the-art NLP tools. Preliminary experimental results suggest that a combination of domain-specific knowledge and information extraction techniques can be used to generalize from the examples and derive more powerful features.

AI and law: a fruitful synergy

AI and Law is a classic field for AI research: it poses difficult and interesting problems for AI, and its projects inform both AI and its focal domain, the law itself. This special issue reports on a range of projects, from those where the law motivates fundamental research and whose results reach beyond the legal domain, to those that partake of the benefits of techniques and wisdom from AI as a whole. For instance, projects tackling legal argument have not only created programs that produce legal arguments but also led to insights and advances in the logic of argumentation. Projects with an applications bent have often provided insights about the limitations and nuances of existing techniques, and have served as the catalysts for developing new approaches. For instance, harnessing models of legal argument to teach law students how to argue has led to refinements to and extensions of the models. There is a synergy not only between law and AI, but also between AI and AI and Law. In fact, the work on Case-Based Reasoning (CBR) done in the AI and Law community provided one of the most important streams of results that contributed to the birth of that area in the mid-1980s. Currently, work on legal argumentation is having a similar impact on the international non-standard logic and argumentation communities. AI and Law is much more than an applications area. Its concerns touch upon issues at the very heart of AI: reasoning, representation, and learning. For the AI researcher interested in symbolic methods—or methods of whatever stripe—that are focused on providing explanations and justifications, AI and Law is an excellent arena. No matter how a reasoner arrives at a legal answer it must be explained, justified, compared to and contrasted with alternatives. For the researcher interested in topics like negotiation, decision-making, e-commerce, natural language, information retrieval and extraction, and data mining, AI and Law is a rich source of problems and inspiration.

Evaluating a learning environment for case-based argumentation skills

CAT0 is an intelligent learning environment designed to help &@ting law students learn basic skills of making arguments with cases, through practice in theory-testing and argumentation tasks. CAT0 models ways in which experts compare and contrast cases, assess the significance of similarities and differences between cases in light of general domain knowledge, and use the same general knowledge to organize multi-case arguments by issues. CAT0 communicates its model to students by presenting dynamically-generated argumentation examples and reifying (i.e., making visible) argument structure. Also, the CAT0 Tools reduce some of the distracting complexity of the students’ task. gal education seeks to teach. But they are difficult to learn and time-consuming to teach. If computer-based instruction can help students get some extra practice in basic skills and free up instructor’s time for more advanced topics, much would be gained. We evaluated CAT0 in the context of a second-semester Iegal writing course taught at the University of Pittsburgh School of Law. We found that 7.5 hours of CAT0 instruction led to a statistically significant improvement in students’ basic argumentation skills, comparable to that achieved, in the same amount of time, by an experienced legal writing instructor teaching small groups of students in a more traditional way. On a more advanced memo-writing assignment, meant to explore the “frontier” of the CAT0 instruction, students taught by the legal writing instructor did better, indicating that more is needed if CAT0 is to help students to improve their memo-writing skills. CAT0 employs a computational model of case-based argumentation that addresses eight basic argument moves and more elaborate multi-case arguments. The model includes a “Factor Hierarchy,” which represents more nbstract., but still domain-specific, legal knowledge about the meaning of the factors used to represent cases. CAT0 uses the Factor Hierarchy for a number of purposes, among them to organize multi-case arguments by issues and to make arguments about the significance of distinctions. To generate the latter, CAT0 strategically selects alternative interpretations of cases to elaborate a “deeper” (or more. abstract) contrast or parallel between cases.

Toward an intelligent tutoring system for teaching law students to argue with cases

This paper describes a research project to devise and test an intelligent, case-baaed tutorial program for teaching law students to argue with cases. In order to present pedagogically interesting lessons and develop a Student Model, we have designed memory structures such as Argument Contexts and a hierarchy of Issues in Case-Baaed Legal Reaaoning. Using logical expressions in the knowledge representation language Loom, we also explicitly represent case-based argument concepts such as a case’s being on point to a problem, more on point than another case, most on point of all the cases, a best case to cite, and a counterexample to another case. The program will be able to reason with the explicit concepts in selecting cases from a Case Library, assembling lessons and examples, analyzing student inputs, and in generating explanations and feedback. We hope to demonstrate empirically that, by providing law students a conceptual model of the criteria for selecting and describing precedents that would be useful in an argument, the tutorial program will help them to learn to select and apply cases more efficiently and to make more effective arguments.