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)

The ability to find relevant materials in large document collections is a fundamental component of legal research. The emergence of large machine-readable collections of legal materials has stimulated research aimed at improving the quality of the tools used to access these collections. Important research has been conducted within the traditional information retrieval, the artificial intelligence, and the legal communities with varying degrees of interaction between these groups. This article provides an introduction to text retrieval (...) and surveys the main research related to the retrieval of legal materials. (shrink)

In this work we study, design, and evaluate computational methods to support interpretation of statutory terms. We propose a novel task of discovering sentences for argumentation about the meaning of statutory terms. The task models the analysis of past treatment of statutory terms, an exercise lawyers routinely perform using a combination of manual and computational approaches. We treat the discovery of sentences as a special case of ad hoc document retrieval. The specifics include retrieval of short texts, specialized (...) document types, and, above all, the unique definition of document relevance provided in detailed annotation guidelines. To support our experiments we assembled a data set comprising 42 queries which we plan to release to the public in the near future in order to support further research. Most importantly, we investigate the feasibility of developing a system that responds to a query with a list of sentences that mention the term in a way that is useful for understanding and elaborating its meaning. This is accomplished by a systematic assessment of different features that model the sentences’ usefulness for interpretation. We combine features into a compound measure that accounts for multiple aspects. The definition of the task, the assembly of the data set, and the detailed task analysis provide a solid foundation for employing a learning-to-rank approach. (shrink)

A variety of legal documents are increasingly being made available in electronic format. Automatic Information Search and Retrieval algorithms play a key role in enabling efficient access to such digitized documents. Although keyword-based search is the traditional method used for text retrieval, they perform poorly when literal term matching is done for query processing, due to synonymy and ambivalence of words. To overcome these drawbacks, an ontological framework to enhance the user’s query for retrieval of (...) truly relevant legal judgments has been proposed in this paper. Ontologies ensure efficient retrieval by enabling inferences based on domain knowledge, which is gathered during the construction of the knowledge base. Empirical results demonstrate that ontology-based searches generate significantly better results than traditional search methods. (shrink)

Legal text retrieval serves as a key component in a wide range of legal text processing tasks such as legal question answering, legal case entailment, and statute law retrieval. The performance of legal text retrieval depends, to a large extent, on the representation of text, both query and legal documents. Based on good representations, a legal text retrieval model can effectively match the query to (...) its relevant documents. Because legal documents often contain long articles and only some parts are relevant to queries, it is quite a challenge for existing models to represent such documents. In this paper, we study the use of attentive neural network-based text representation for statute law document retrieval. We propose a general approach using deep neural networks with attention mechanisms. Based on it, we develop two hierarchical architectures with sparse attention to represent long sentences and articles, and we name them Attentive CNN and Paraformer. The methods are evaluated on datasets of different sizes and characteristics in English, Japanese, and Vietnamese. Experimental results show that: (i) Attentive neural methods substantially outperform non-neural methods in terms of retrieval performance across datasets and languages; (ii) Pretrained transformer-based models achieve better accuracy on small datasets at the cost of high computational complexity while lighter weight Attentive CNN achieves better accuracy on large datasets; and (iii) Our proposed Paraformer outperforms state-of-the-art methods on COLIEE dataset, achieving the highest recall and F2 scores in the top-N retrieval task. (shrink)

Legal professionals in Malawi rely on a limited number of textbooks, outdated law reports and inadequate library services. Most documents available are in image form, are un-structured, i.e. contain no useful legal meta-data, summaries, keynotes, and do not support a system of citation that is essential to legal research. While advances in document processing and machine learning have benefited many fields, legal research is still only marginally affected. In this interdisciplinary research, the authors build semi-automatic tools (...) for creating a corpus of Malawi criminal law decisions annotated with legal meta-data, case and law citations. We used this corpus to extract legal meta-data, including law and case citations as used in Malawi by employing machine learning tools, spaCy and Gensim LDA. We set the foundation for a new methodology for classifying Malawi criminal case law according to the recently introduced International Classification of Crime for Statistical Purposes (ICCS). (shrink)

A large interest has been dedicated in recent years to the study of models for textual databases amenable to an effective integration of search and navigation functions. In the field of legal databases the need for sophisticated models is emphasised by the need to relate and combine in an effective way different types of texts, in order to solve legal problems.In our research we have analysed several existing models, each providing specific benefits and exhibiting corresponding limitations, under both (...) a functional and economical viewpoint. (shrink)

Legal Judgment Summarization (LJS) can highly summarize legal judgment documents, improving judicial work efficiency in case retrieval and other occasions. Legal judgment documents are usually lengthy; however, most existing LJS methods are directly based on general text summarization models, which cannot handle long texts effectively. Additionally, due to the complex structural characteristics of legal judgment documents, some information may be lost by applying only one single kind of summarization model. To address these issues, we (...) propose an integrated summarization method which leverages both semantic and structural information to improve the quality of LJS. Specifically, legal judgment documents are firstly segmented into three relatively short parts according to their specific structure. We propose an extractive summarization model named BSLT and an abstractive summarization model named LPGN by adopting Lawformer as the encoder. Lawformer is a new pre-trained language model for long legal documents, which specializes in capturing long-distance dependency and modeling legal semantic features. Then, we adopt different models to summarize the corresponding part regarding its structural characteristics. Finally, the obtained summaries are integrated to generate a high-quality summary involving semantic and structural information. We conduct comparative experiments to evaluate the performance of our model. The results show that our model outperforms the baseline model LEAD-3 by 14.78% on the mean ROUGE score, which demonstrates our method is effective in LJS and is prospected to be applied to assist other tasks in legal artificial intelligence. (shrink)

The effectiveness of information retrieval technology in electronic discovery (E-discovery) has become the subject of judicial rulings and practitioner controversy. The scale and nature of E-discovery tasks, however, has pushed traditional information retrieval evaluation approaches to their limits. This paper reviews the legal and operational context of E-discovery and the approaches to evaluating search technology that have evolved in the research community. It then describes a multi-year effort carried out as part of the Text Retrieval (...) Conference to develop evaluation methods for responsive review tasks in E-discovery. This work has led to new approaches to measuring effectiveness in both batch and interactive frameworks, large data sets, and some surprising results for the recall and precision of Boolean and statistical information retrieval methods. The paper concludes by offering some thoughts about future research in both the legal and technical communities toward the goal of reliable, effective use of information retrieval in E-discovery. (shrink)

Case law retrieval is the task of locating truly relevant legal cases given an input query case. Unlike information retrieval for general texts, this task is more complex with two phases (legal case retrieval and legal case entailment) and much harder due to a number of reasons. First, both the query and candidate cases are long documents consisting of several paragraphs. This makes it difficult to model with representation learning that usually has restriction on (...) input length. Second, the concept of relevancy in this domain is defined based on the legal relation that goes beyond the lexical or topical relevance. This is a real challenge because normal text matching will not work. Third, building a large and accurate legal case dataset requires a lot of effort and expertise. This is obviously an obstacle to creating enough data for training deep retrieval models. In this paper, we propose a novel approach called supporting model that can deal with both phases. The underlying idea is the case–case supporting relation and the paragraph–paragraph as well as the decision-paragraph matching strategy. In addition, we propose a method to automatically create a large weak-labeling dataset to overcome the lack of data. The experiments showed that our solution has achieved the state-of-the-art results for both case retrieval and case entailment phases. (shrink)

The paper provides an OWL ontology for legal cases with an instantiation of the legal case Popov v. Hayashi. The ontology makes explicit the conceptual knowledge of the legal case domain, supports reasoning about the domain, and can be used to annotate the text of cases, which in turn can be used to populate the ontology. A populated ontology is a case base which can be used for information retrieval, information extraction, and case based reasoning. (...) The ontology contains not only elements for indexing the case (e.g. the parties, jurisdiction, and date), but as well elements used to reason to a decision such as argument schemes and the components input to the schemes. We use the Protégé ontology editor and knowledge acquisition system, current guidelines for ontology development, and tools for visual and linguistic presentation of the ontology. (shrink)

In this work, we provide a broad overview of the distinct stages of E-Discovery. We portray them as an interconnected, often complex workflow process, while relating them to the general Electronic Discovery Reference Model (EDRM). We start with the definition of E-Discovery. We then describe the very positive role that NIST’s Text REtrieval Conference (TREC) has added to the science of E-Discovery, in terms of the tasks involved and the evaluation of the legal discovery work performed. Given (...) the critical nature that data analysis plays at various stages of the process, we present a pyramid model, which complements the EDRM model: for gathering and hosting; indexing; searching and navigating; and finally consolidating and summarizing E-Discovery findings. Next we discuss where the current areas of need and areas of growth appear to be, using one of the field’s most authoritative surveys of providers and consumers of E-Discovery products and services. We subsequently address some areas of Artificial Intelligence, both Information Retrieval-related and not, which promise to make future contributions to the E-Discovery discipline. Some of these areas include data mining applied to e-mail and social networks, classification and machine learning, and the technologies that will enable next generation E-Discovery. The lesson we convey is that the more IR researchers and others understand the broader context of E-Discovery, including the stages that occur before and after primary search, the greater will be the prospects for broader solutions, creative optimizations and synergies yet to be tapped. (shrink)

Sentence boundary detection (SBD) represents an important first step in natural language processing since accurately identifying sentence boundaries significantly impacts downstream applications. Nevertheless, detecting sentence boundaries within legal texts poses a unique and challenging problem due to their distinct structural and linguistic features. Our approach utilizes deep learning models to leverage delimiter and surrounding context information as input, enabling precise detection of sentence boundaries in English legal texts. We evaluate various deep learning models, including domain-specific transformer models like (...) LegalBERT and CaseLawBERT. To assess the efficacy of our deep learning models, we compare them with a state-of-the-art domain-specific statistical conditional random field (CRF) model. After considering model size, F1-score, and inference time, we identify the Convolutional Neural Network Model (CNN) as the top-performing deep learning model. To further enhance performance, we integrate the features of the CNN model into the subsequent CRF model, creating a hybrid architecture that combines the strengths of both models. Our experiments demonstrate that the hybrid model outperforms the baseline model, achieving a 4% improvement in the F1-score. Additional experiments showcase the superiority of the hybrid model over SBD open-source libraries when confronted with an out-of-domain test set. These findings underscore the importance of efficient SBD in legal texts and emphasize the advantages of employing deep learning models and hybrid architectures to achieve optimal performance. (shrink)

Chapter 1: An Introduction: The ‘Semantic Sting’ Argument Describes Dworkin’s theory as concerning the conditions of legal validity. “A legal system is a system of norms. Validity is a logical property of norms in a way akin to that in which truth is a logical property of propositions. A statement about the law is true if and only if the norm it purports to describe is a valid legal norm…It follows that there must be certain conditions which (...) render certain norms, but not others, legally valid. Hence it also follows that there can be two types of disagreement over the truth or falsehood of propositions about the law. People can disagree over the question ‘What are the conditions of legal validity?’, in which case their disagreement is a theoretical one. Or they can agree on the conditions of validity, and disagree as to whether or not those conditions actually obtain in a given case or not.” Argues that the best way to interpret Dworkin’s ‘semantic-sting argument’ is as a general argument against conventionalism in legal positivism. Summarizes the outcome of Dworkin’s argument as follows: “[I]f Dworkin is right about the legal reasoning of lawyers and judges, conventionalism would be self-defeating: if lawyers and judges recognize as legally binding not only those norms which are uncontroversially identifiable under the Rule of Recognition, that is, if what they recognize as binding is not only source-based law, then conventionalism turns out to be false on its own terms. In other words, either law is not what lawyers and judges think that it is, in which case law is not a matter of conventions, or – if it is what lawyers and judges think – conventionalism is false, as they do not see the law as purely a matter of conventions.” Dworkin’s theory, then, is a new conception of jurisprudence aiming to present itself as a rival to conventionalism. The book will be concerned with re-examining positivism in light of the interpretative challenge. Chapter 2: Meaning and Interpretation “[R]oughly, interpretation can be defined as an understanding or explanation of the meaning of an object.” Interpretation in this narrow sense is distinct from a less formal and broader use of interpretation that makes it equivalent to explanation. “Hence also, only those objects which are capable of bearing some meaning qualify as objects of interpretation.” [Why shouldn’t we think of interpretation as object-identification? This seems to be closer to what Dworkin has in mind.] Thus, a consideration of the philosophical work on meaning is appropriate. First, a consideration of Davidson on interpretation. The thesis is that Davidson’s “radical interpretation” can only account for sentence meaning and not for linguistic interpretation in general. Davidson cannot account for the non-rule-governed aspects of communication. “Semantics, as opposed to interpretation, concerns those aspects of communication which are rule or convention governed…but such rules are normally unavailable as reasons or justifications for an interpretation. On the contrary, interpretation is usually required because the issue is not determined by rules or conventions.” Interpretation, rather, is parasitic with understanding the meaning of an expression and not equivalent with it. As Dummett notes, any reflection on the meanings of words assumes prior knowledge of ordinary meanings. “This leads to the conclusion that understanding or explaining the meaning of an expression and interpreting it, are two conceptually separate things. It also indicates that seman ics can only be employed, if at all, to elucidate the concept of interpretation by way of contrast: interpretation concerns those aspects of communication which are under-determined by semantic rules or conventions.” Pragmatics, the study of problems posed by discrepancies between utterer’s meaning and sentence meaning, would seem to be closer to inquiry into interpretation. Both are concerned with understanding how meaning is possible where conventional rules leave off. Contemporary pragmatics, moreover, recognizes the logically indeterminate character of communicative inferences beyond conventional meaning and that the contextual conditions that account for the possibility cannot be realized in the semantic structure of the sentence. However, different criteria of success may be at work in pragmatics than interpretation. Pragmatics is concerned with how communication is achieved and thus the criteria of success would be grasping a speaker’s intentions. Many think, however, that communication is unimportant in certain spheres of communication, e.g. artistic interpretation, and that in these cases the aim of interpretation is not to uncover the actual artist’s intentions. What kind of meaning is interpretation aimed at then? It cannot be the meaning the object has for the interpreter. “Interpretation purports to be a statement about the object interpreted, not about the subject who offers the interpretation…it is crucial to remember that not everything we say about a work of art, or any other text, amounts to an interpretation of it. Only those aspects of a text which can shed light on its meaning form part of the text’s interpretation. What the text means for someone rarely entails anything about what the text means.” [This is puzzling to some extent for it remains somewhat ambiguous what it means to talk of subject-independent meaning.] Interpretive meaning attribution is, rather, counterfactual. “In general, I will suggest that the answer to this question consists in the fact that meaning is assigned by a counterfactual statement. Given that x is the meaning attributed to, for instance a text T, and x is not the literal meaning of T, nor is it the meaning of T intended by its author, then the attribution of meaning x to T can only be understood as the contention that on the basis of certain assumptions a certain fictitious or stipulated speaker would have meant x by expressing T.” Thus, on this account, interpretation, when it does not concern an actual speaker, turns out to be a kind of counterfactual intention attribution. Twofold point: “First, that interpretation is essentially a matter of attributing intentions, that is, in the pragmatics sense of ‘meaning’, namely, meaning that such-and-such by an act or expression. At the same time, interpretations need not be based on the intentions of actual authors; the meaning of an act or expression is understandable in terms of counterfactual intentions.” [Are we really restricted to understanding the coherence and integrity of meaning in such a way? Presumably the author is a kind of device for organizing meaning, much like Dworkin’s integrity, but need it take that format?] The logic of interpretation is typically reducible to intention attribution. [What role does the counterfactual author play in structuring meaning?] The criteria of interpretive success will depend upon the object, whether it is the real author or the counterfactual author. If the former, success will depend upon the retrieval of the author’s actual intentions. “Likewise, if the author is characterized in terms of some ideal representative of a certain genre, for instance, the presumptions which are taken to determine this c aracterization would provide the criteria of success for the particular interpretation offered.” [The criteria of success would seem, then, to be the principles, ideals, etc. that we attribute to the counterfactual author. These attributed qualities would be doing all the work – is the attribution of intention to the hypothetical author, then, really necessary at all?] [Does this account presuppose a weird dichotomy between what an object is and what it means? This dichotomy is different, if present at all, in Dworkin.] Chapter 3: Dworkin’s Theory of Interpretation and the Nature of Jurisprudence Three main insights of Dworkin’s theory of interpretation: “First, that interpretation strives to present its object in its best possible light. Second, that interpretation is essentially genre-dependent. And third, that there are certain constraints that determine the limits of possible interpretations of a given object.” Why the best light? Dworkin doesn’t really address this question in depth but claims that otherwise we are left with no way of justifying and choosing an interpretation or we are left with the author’s intention model, which we should reject. [Marmor doesn’t seem clear on what precisely it means to present something in its best light – seems to conflate it with most appealing light which, I take it, is different than what Dworkin has in mind.] “Dworkin is quite right to maintain that without having some views about the values inherent in the genre to which the text is taken to belong, no interpretation can take off the ground. The values we associate with the genre partly, but crucially, determine what would make sense to say about the text, what are the kinds of meaning we could ascribe to it.” There is no necessity for an interpretation to present an object in its best light with regards to the genre it is taken to be part of. Interpretations “could simply strive to present it in a certain light, perhaps better than some, worse than others, but in a way which highlights an aspect of the meaning of the text which may be worth paying attention to for some reason or another.” [What kind of reasons would count in choosing to present it in one light rather than another?] There is simply zero grounds for Dworkin’s assertion that we have no reason to pay attention to an interpretation that does not present it in its best light. Take a psychoanalytic interpretation of Hamlet, for example – not the best light, but a legitimate one. [This seems related to the next criticism – clearly Dworkin would have to loosen his insistence on the best, though not to detriment.] Moreover, especially with regards to art, it doesn’t appear possible to make an all things considered judgment about the best interpretation given the incommensurability of different values that are taken to inhere in a given genre. [Dworkin can respond here that we may have different available interpretations that could be justified with respect to the relevant value. Moreover, a lack of common denominator does not prevent any comparison whatsoever.] Constructive interpretation concerns social practices constituted by norms – it imposes a purpose or value rendering intelligible the normative character of the practice. There is a separate aspect of constructive interpretation that maintains that elements of the practice are sensitive to the point of the practice. It is important to note that different social practices will institutionalize themselves in different ways – thus, different practices will establish varying conditions on how the practice can be altered. The position of the positivists is that evaluative judgments concerning what the law should be are insufficient for determining what the law is given the nature of law’s institutionalization.&nb p; [This would seem to reside on a certain conception of law, i.e. an interpretation about where the value of law lies.] “[T]his being one of the main points of dispute between Dworkin and his positivist opponents, Dworkin cannot at this initial state presume law to be sensitive to its value in the manner that other, non-institutional practices might be, without incurring the charge of having assumed the very point at issue.” The upshot for Dworkin is that any explanation of a social practice such as law requires the same kind of reasoning as participation in that practice. This will be called the hermeneutic thesis. We can try to reconstruct a justification for this view based on the necessity of shared background assumptions – but this will not do since Dworkin’s claim is stronger: it contends that participants’ and theorists’ must “adopt one and the same normative point of view.” [The sense of normative here is unclear. It is true, though, that shared background assumptions do not imply viewing the law as valuable in any particular way, or at least it does not imply this in any strong sense.] Dworkin makes two distinct points with the hermeneutic thesis: “First, that the explanation of a social practice, like law or the arts, is essentially interpretative, and as such, necessarily value laden. Second, that the interpretation of such social practices, which Dworkin calls ‘argumentative’, is unique in the sense that the practice itself is an evaluative enterprise, and that therefore the interpreter of such a practice must form an evaluative judgment of her own about those values which are inherent in the practice that she purports to interpret.” [As stated, the first thesis is terribly ambiguous – for Dworkin, it clear implies a commitment to constructive interpretation. For Marmor, who accepts it, it is totally unclear what commitment he intends to make. He seems to mean simply that we have criteria for success concerning what counts as a successful explanation.] Mamor denies the second thesis. “What Dworkin seems to ignore here is that there is a crucial difference between forming a view about the values which are manifest in a social practice, like law, and actually having evaluative judgments about them.” [The problem, however, it that it is a contentious matter what values are part of the law – a theorist’s account of what values are constitutive of a particular legal system must rely on the kind of arguments which participants rely on. Of course, a theorist might simply note a debate about values and attempt to remain agnostic, but this will fail to be a full theoretical account of what law is!] Dworkin’s thesis seems to depend exclusively on his account of constructive interpretation. [This seems exactly right.] Chapter 4: Coherence, Holism, and Interpretation: The Epistemic Foundation of Dworkin’s Legal Theory Begins with discussion of Rawls’ reflective equilibrium. Sums up dual criticism as follows: “[T]he assumption that intuitions are independently true and the converse one, that their truth depends on fitting a coherent scheme, both seem to yield paradoxical results.” Dworkin, in his article “The Original Position”, claims that Rawls cannot commit himself to ethical realism. “[T]he natural model presupposes some form of ethical realism, while the constructive model does not. One important difficulty arising from an attempt to apply the natural model to Rawls is, that under the natural model, any theory which dopes not account for an intuition, at least for one which is held firmly, cannot be wholly satisfactory, just as scientific theory which does not account for certain observational data it is supposed to cover, would not be satisfactory in a familiar way.” Rawls, rather, is committed to a constructivist view of morality. Moreover, the reason we value coherence epistemically in moral theory is that it itself is a value of political morality – fairness requires consistency and publicity in the application of moral standards. Marmor notes the following problem with such a view: “[I]f coherence is justified, as it is here, with reference to certain moral values, that is, a specific conception of fairness, then we face the following problem: the presupposed values of fairness must themselves be based upon intuitive convictions, in which case the question of their truth cannot be ignored. If they are taken to be true…we are driven back to the perplexities of the natural model.” Discusses, helpfully, the Fish/Dworkin debate. The upshot is that Dworkin’s interpretive model is best off if it commits itself to a Quineian holism. However, this undermines his distinction between internal and external skepticism since holism must deny that moral judgments constitute a closed system. [This analysis seems exactly right.] Endorses Simmonds view that Dworkin’s theory does not meet the requirements of complexity he identifies as necessary to avoid vicious circularity in interpretive judgments of fit and identity. Coherence is doing the work through and through and therefore the interpretive theory of law is circular. [Will require further investigation, though it is unclear how coherence is doing any work in the pre-interpretive stage. At this stage, the judgments, if interpretive, seem to be derived from different sources. In any case, using coherence as a value seems to be highly suspect.] Why doesn’t Dworkin just reject Fish’s assumptions: because of his jurisprudence. “If legal texts can have a meaning that is not entirely dependent on a process of interpretation, then it is at least sometimes the case that the law can simply be understood, and applied, without the mediation of interpretation. And if that is the case, then the argument from interpretation against legal positivism collapses. It is no longer the case that every conclusion about what the law is, depends on evaluative considerations about what it ought to be.” Chapter 5: Semantics, Realism, and Natural Law Assessment of Michael Moore’s legal realism and its implications. [The final criticisms of Moore are not decisive.] Chapter 6: Constructive Identification and Razian Authority A consideration of Dworkin’s denial that the communication model of interpretation is appropriate for legal interpretation. Marmor argues that intentions do play a crucial role in the identification of legal norms in a way that is incompatible with Dworkin’s “coherence thesis”. Marmor takes Dworkin to be committed to the constructive identification thesis: that the identification of something as part of the legal, artistic, etc. genre can be done sufficiently by evaluative considerations. “Here, one must maintain that evaluative considerations are sufficient to determine whether something is a legal norm or a work of art”. [Fit, however, constrains possible evaluative considerations – thus evaluative considerations are not sufficient in themselves.] “Now the crucial point here is this: if you maintain the possibility of constructive identification in art, you must assume that works of art can be identified as such on the basis of certain features they happen to possess, features which contain no reference to any particular intention to create a work of art…unless we take intentions into account, how can we discriminate between the concept of an aesthetic artifact and the con ept of a work of art?” [It is unclear, and I think false, that constructive interpretation of anything requires identifying something irrespective of an artificers intentions. Those intentions may be relevant, dependent upon the genre under which we are trying to classify the object. Nothing, seemingly, about constructive interpretation in general excludes this possibility.] Marmor’s point is that it turns out to be impossible to identify art as art without reference to an artist’s intentions – especially given the state of contemporary art. [It is notable that Marmor’s response to counterexample implicitly depends upon a fully developed theory of aesthetics. His cursory comments on the matter of identifying art seem insufficient to say the least.] The rest of the chapter considers the incompatibility of Raz’s account of authority with constructive interpretation. The incompatibilities mentioned do not move beyond what Raz identifies in his article “Authority, Law and Morality”, but a defense of this notion of authority is defended in several [inconclusive] respects. [Raz and Marmor want to separate identification from content. This is not clearly a defensible separation.] Chapter 7: No Easy Cases? Chapter defends notion that there is a distinction between easy and hard cases in the positivist sense. Takes a Hartian approach and ultimately defends the distinction by associating it Wittgenstein’s conception of rule following. The point is that one need to look the purpose of rule in order to understand what the rule requires. Chapter 8: Legislative Intent and the Authority of Law Examines the following doctrine concerning the role of legislative intent in adjudication: “[F]irst, it would hold that laws, at least in certain cases, are enacted with relatively specific intentions, and that this is a matter of fact which is discernible through an ordinary fact-finding procedure. Second, that in certain cases the presence of such a fact, namely, that the law was enacted with a certain intention, provides judges with a reason to decide the legal dispute in accordance with the relevant legislative intent.” [It is important that for Marmor, interpretive strategies are only appropriate in hard cases. However, we should wonder what grounds positivism can offer for deciding a hard case in any manner which is not the morally best manner according to the judgment of the judge. There are no legal grounds, by definition, in hard cases and so what justificatory strategy would endorse any judicial decision except for the morally best one?] We can ascribe intentions to a legislative body, when this is possible at all, by taking the shared intentions of the majority, rather than some group intention, to be the significant intentions. Moreover, we should expect a good deal of consensus about intentions for otherwise it is hard to explain how legislative bodies are able to produce so much legislation. Discussion of kinds of intentions concludes as follows: “I have distinguished between three main types of intention that are potentially relevant from the legal point of view. Apart from the intentions that are manifest in the language of the law itself, legislators typically have further intentions in enacting a given law, and sometimes they would have certain intentions bearing on its proper application. I have also suggested that some of these further intentions may be essentially non-avowable, in which case they are rendered initially irrelevant. Finally, I have pointed out that considerations of consistency require that the legislator’s application intentions be taken into account only if, and to the extent that, they are in accord with his further intentions.” The manner in which intentionalism is to be justif ed relies on the distinction, Marmor maintains, between expert and mere collective action authority. “The point I wanted to make is strictly conditional: if, and only if, a certain law is justified on the basis of the expertise branch of the normal justification thesis, would it make sense to defer to the legislature’s intentions in the interpretation of the law, that is, to the extent that there is, in fact,; such an intention and it can clarify something that needs clarification.” Insofar as a legislature can be considered an expert on the matter requiring adjudication and interpretation should the body’s intention be taken into account. Chapter 9: Constitutional Interpretation Begins by laying out general necessary features of a constitution. Then Marmor moves on to consider general questions of constitutional legitimacy. First, he considers what grounds might be offered for the legitimacy of a constitution at all. He notes that the central question is what permits a single generation to bind future generations. He then examines four arguments to avoid this problem. The first is based on the moral legitimacy of the constitution; Marmor rejects this solution for the [truly dubious] reason that the constitution would make no practical difference because it would not supply reasons in addition to those provided by morality [but, presumably, the point is that these legally valid rules would be institutionally enshrined]. The second is based on the moral authority of the framers and Marmor rightly rejects this on the basis of the idealization of the framers it requires and less rightly on the basis that there cannot be moral authorities. Third is the “argument from interpretation” which claims that “as long as the particular content of the constitution is determined by its interpretation, and the authoritative interpretation at any time correctly instantiates the values which ought to be upheld in the community, the constitution would be morally legitimate”. The fourth is Raz’s which argues that a constitution is valid so long as its constraints are morally permissible because conventions of this type might aid the continuity of the legal system, i.e. it is important to have a convention of this type and the constitution is a kind of this type. Marmor appears to endorse these two approaches and concludes: “The conditions for the legitimacy of a constitution must comprise the following conditions. First, the values and principles enshrined in it must be morally permissible…Second, when certain choices are made in particular cases, they would be legitimate if they are either morally underdetermined, or else, morally correct…It follows from this that both arguments must assume that at least in those areas in which the constitution would make a moral difference, it can be interpreted to make the difference that it should, that is, according to the true moral principles that should apply to the particular case...the moral legitimacy of constitutions very much dependent on the practices of their interpretation. In other words, a great deal of the burden of moral legitimacy is shifted by these arguments to the application of the constitution, thus assuming that the constitution is legitimate only if the courts are likely to apply the constitution in a morally desirable way.” [This implies that the only significant value constitutions have is in providing a conventional practice where some such practice is required.] Moves on to argue that courts typically rely, and rightly so, on moral considerations when deciding constitutional issues because typically, in cases of the kind that reach the supreme or constitutional court, there is no law on the case. Much of this relies on his earlier discussion about the nature of interpretation in the law – that legal interpretation is always a matter of cha ging the law. (shrink)

In the domain of legal information retrieval, an important challenge is to compute similarity between two legal documents. Precedents play an important role in The Common Law system, where lawyers need to frequently refer to relevant prior cases. Measuring document similarity is one of the most crucial aspects of any document retrieval system which decides the speed, scalability and accuracy of the system. Text-based and network-based methods for computing similarity among case reports have already been (...) proposed in prior works but not without a few pitfalls. Since legal citation networks are generally highly disconnected, network based metrics are not suited for them. Till date, only a few text-based and predominant embedding based methods have been employed, for instance, TF-IDF based approaches, Word2Vec and Doc2Vec based approaches. We investigate the performance of 56 different methodologies for computing textual similarity across court case statements when applied on a dataset of Indian Supreme Court Cases. Among the 56 different methods, thirty are adaptations of existing methods and twenty-six are our proposed methods. The methods studied include models such as BERT and Law2Vec. It is observed that the more traditional methods that rely on a bag-of-words representation performs better than the more advanced context-aware methods for computing document-level similarity. Finally we nominate, via empirical validation, five of our best performing methods as appropriate for measuring similarity between case reports. Among these five, two are adaptations of existing methods and the other three are our proposed methods. (shrink)

There is more to legal knowledge representation than knowledge-bases. It is valuable to look at legal knowledge representation and its implementation across the entire domain of computerisation of law, rather than focussing on sub-domains such as legal expert systems. The DataLex WorkStation software and applications developed using it are used to provide examples. Effective integration of inferencing, hypertext and text retrieval can overcome some of the limitations of these current paradigms of legal computerisation which (...) are apparent when they are used on a stand-alone basis. Effective integration of inferencing systems is facilitated by use of a (quasi) natural language knowledge representation, and the benefits of isomorphism are enhanced. These advantages of integration apply to all forms of inferencing, including document generation and casebased inferencing. Some principles for development of integrated legal decision support systems are proposed. (shrink)

The standard approach for abstractive text summarization is to use an encoder-decoder architecture. The encoder is responsible for capturing the general meaning from the source text, and the decoder is in charge of generating the final text summary. While this approach can compose summaries that resemble human writing, some may contain unrelated or unfaithful information. This problem is called “hallucination” and it represents a serious issue in legal texts as legal practitioners rely on these summaries (...) when looking for precedents, used to support legal arguments. Another concern is that legal documents tend to be very long and may not be fed entirely to the encoder. We propose our method called LegalSumm for addressing these issues by creating different “views” over the source text, training summarization models to generate independent versions of summaries, and applying entailment module to judge how faithful these candidate summaries are with respect to the source text. We show that the proposed approach can select candidate summaries that improve ROUGE scores in all metrics evaluated. (shrink)

Similar case analysis (SCA) is an essential topic in legal artificial intelligence, serving as a reference for legal professionals. Most existing works treat SCA as a traditional text classification task and ignore some important legal elements that affect the verdict and case similarity, like legal events, and thus are easily misled by semantic structure. To address this issue, we propose a Legal Event-Context Model named LECM to improve the accuracy and interpretability of SCA based (...) on Chinese legal corpus. The event-context integration mechanism, which is an essential component of the LECM, is proposed to integrate the legal event and context information based on the attention mechanism, enabling legal events to be associated with their corresponding relevant contexts. We introduce an event detection module to obtain the legal event information, which is pre-trained on a legal event detection dataset to avoid labeling events manually. We conduct extensive experiments on two SCA tasks, i.e., similar case matching (SCM) and similar case retrieval (SCR). Compared with baseline models, LECM is validated by about 13% and 11% average improvement in terms of mean average precision and accuracy respectively, for SCR and SCM tasks. These results indicate that LECM effectively utilizes event-context knowledge to enhance SCA performance and its potential application in various legal document analysis tasks. (shrink)

Natural language processing techniques contribute more and more in analyzing legal documents recently, which supports the implementation of laws and rules using computers. Previous approaches in representing a legal sentence often based on logical patterns that illustrate the relations between concepts in the sentence, often consist of multiple words. Those representations cause the lack of semantic information at the word level. In our work, we aim to tackle such shortcomings by representing legal texts in the form of (...) abstract meaning representation, a graph-based semantic representation that gains lots of polarity in NLP community recently. We present our study in AMR Parsing and AMR-to-text Generation specifically for legal domain. We also introduce JCivilCode, a human-annotated legal AMR dataset which was created and verified by a group of linguistic and legal experts. We conduct an empirical evaluation of various approaches in parsing and generating AMR on our own dataset and show the current challenges. Based on our observation, we propose our domain adaptation method applying in the training phase and decoding phase of a neural AMR-to-text generation model. Our method improves the quality of text generated from AMR graph compared to the baseline model. 2018; and “Legal Text Generation fromMeaning Representation”, published in the 32nd International Conference on Legal Knowledge and Information Systems 2019.). (shrink)

The standard approach for abstractive text summarization is to use an encoder-decoder architecture. The encoder is responsible for capturing the general meaning from the source text, and the decoder is in charge of generating the final text summary. While this approach can compose summaries that resemble human writing, some may contain unrelated or unfaithful information. This problem is called “hallucination” and it represents a serious issue in legal texts as legal practitioners rely on these summaries (...) when looking for precedents, used to support legal arguments. Another concern is that legal documents tend to be very long and may not be fed entirely to the encoder. We propose our method called LegalSumm for addressing these issues by creating different “views” over the source text, training summarization models to generate independent versions of summaries, and applying entailment module to judge how faithful these candidate summaries are with respect to the source text. We show that the proposed approach can select candidate summaries that improve ROUGE scores in all metrics evaluated. (shrink)

A large number of cross-references to various bodies of text are used in legal texts, each serving a different purpose. It is often necessary for authorities and companies to look into certain types of these citations. Yet, there is a lack of automatic tools to aid in this process. Recently, citation graphs have been used to improve the intelligibility of complex rule frameworks. We propose an algorithm that builds the citation graph from a document and automatically labels each (...) edge according to its purpose. Our method uses the citing text only and thus works only on citations who’s purpose can be uniquely identified by their surrounding text. This framework is then applied to the US code. This paper includes defining and evaluating a standard gold set of labels that cover a vast majority of citation types which appear in the “US Code” but are still short enough for practical use. We also proposed a novel linear-chain conditional random field model that extracts the features required for labeling the citations from the surrounding text. We then analyzed the effectiveness of different clustering methods such as K-means and support vector machine to automatically label each citation with the corresponding label. Besides this, we talk about the practical difficulties of this task and give a comparison of human accuracy compared to our end-to-end algorithm. (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 addresses the problems that lawyers experience retrieving information from legal-text databases. Traditional access mechanisms of text databases require users to know how information is stored. We propose a method for index organisation which shields lawyers from the internal storage structures and which allows them to address the legal databases in their own legal terms. The proposed index is based on a model of legal tasks as opposed to traditional database indexes which represent (...) the contents of the database. We will lay out the architecture of an information system in which this task model is used to determine the information need, to retrieve relevant documents and to give methodical guidance for the legal task itself. To account for the design of a task-based legal information retrieval system, a substantial part of this paper is devoted to analysis and representation of legal tasks. (shrink)

The SALOMON project is a contribution to the automatic processing of legal texts. Its aim is to automatically summarise Belgian criminal cases in order to improve access to the large number of existing and future cases. Therefore, techniques are developed for identifying and extracting relevant information from the cases. A broader application of these techniques could considerably simplify the work of the legal profession.A double methodology was used when developing SALOMON: the cases are processed by employing additional knowledge (...) to interpret structural patterns and features on the one hand and by way of occurrence statistics of index terms on the other. As a result, SALOMON performs an initial categorisation and structuring of the cases and subsequently extracts the most relevant text units of the alleged offences and of the opinion of the court. The SALOMON techniques do not themselves solve any legal questions, but they do guide the user effectively towards relevant texts. (shrink)

Financial risks are among the most important risks in the construction industry projects, which significantly impact project objectives, including project cost. Besides, financial risks have many interactions with each other and project parameters, which must be taken into account to analyze risks correctly. In addition, a source of financial risks in a project is the contract, which is the most important project document. Identifying terms related to financial risks in a contract and considering their effects on the risk management process (...) is an essential issue that has been neglected. Hence, an integrated model for evaluating financial risks and their related contractual clauses were presented. To this end, the effect of financial risks on the project cost was simulated using a system dynamics model. Moreover, terms related to financial risks in a contract text were identified and extracted using text mining, and their effect was included in the system dynamics model. The model was implemented in a hospital construction project in Tehran as a case study, and its results were analyzed. The innovation of the research is integrating text mining and the system dynamics model to investigate the effect of financial risks and related contractual clauses on the project cost. (shrink)

Named entity recognition (NER) is a very relevant task for text information retrieval in natural language processing (NLP) problems. Most recent state-of-the-art NER methods require humans to annotate and provide useful data for model training. However, using human power to identify, circumscribe and label entities manually can be very expensive in terms of time, money, and effort. This paper investigates the use of prompt-based language models (OpenAI’s GPT-3) and weak supervision in the legal domain. We apply both (...) strategies as alternative approaches to the traditional human-based annotation method, relying on computer power instead human effort for labeling, and subsequently compare model performance between computer and human-generated data. We also introduce combinations of all three mentioned methods (prompt-based, weak supervision, and human annotation), aiming to find ways to maintain high model efficiency and low annotation costs. We showed that, despite human labeling still maintaining better overall performance results, the alternative strategies and their combinations presented themselves as valid options, displaying positive results and similar model scores at lower costs. Final results demonstrate preservation of human-trained models scores averaging 74.0% for GPT-3, 95.6% for weak supervision, 90.7% for GPT + weak supervision combination, and 83.9% for GPT + 30% human-labeling combination. (shrink)

Regulations affect every aspect of our lives. Compliance with the regulations impacts citizens and businesses similarly: they have to find their rights and obligations in the complex legal environment. The situation is more complex when languages and time versions of regulations should be considered. To propose a solution to these demands, we present a semantic enrichment approach which aims at (1) decreasing the ambiguousness of legal texts, (2) increasing the probability of finding the relevant legal materials, and (...) (3) utilizing the application of legal reasoners. Our approach is also implemented both as a service for citizens and businesses and as a modeling environment for legal drafters. To evaluate the usefulness of the approach, a case study was carried out in a large organization and applied to corporate regulations and Hungarian laws. The results suggest this approach can support the previous aims. (shrink)

Exploratory analysis is an area of increasing interest in the computational linguistics arena. Pragmatically speaking, exploratory analysis may be paraphrased as natural language processing by means of analyzing large corpora of text. Concerning the analysis, appropriate means are statistics, on the one hand, and artificial neural networks, on the other hand. As a challenging application area for exploratory analysis of text corpora we may certainly identify text databases, be it information retrieval or information filtering systems. With (...) this paper we present recent findings of exploratory analysis based on both statistical and neural models applied to legal text corpora. Concerning the artificial neural networks, we rely on a model adhering to the unsupervised learning paradigm. This choice appears naturally when taking into account the specific properties of large text corpora where one is faced with the fact that input-output-mappings as required by supervised learning models cannot be provided beforehand to a satisfying extent. This is due to the fact of the highly changing contents of text archives. In a nutshell, artificial neural networks count for their highly robust behavior regarding the parameters for model optimization. In particular, we found statistical classification techniques much more susceptible to minor parameter variations than unsupervised artificial neural networks. In this paper we describe two different lines of research in exploratory analysis. First, we use the classification methods for concept analysis. The general goal is to uncover different meanings of one and the same natural language concept. A task that, obviously, is of specific importance during the creation of thesauri. As a convenient environment to present the results we selected the legal term of neutrality, which is a perfect representative of a concept having a number of highly divergent meanings. Second, we describe the classification methods in the setting of document classification. The ultimate goal in such an application is to uncover semantic similarities of various text documents in order to increase the efficiency of an information retrieval system. In this sense, document classification has its fixed position in information retrieval research from the very beginning. Nowadays renewed massive interest in document classification may be witnessed due to the appearance of large-scale digital libraries. (shrink)

Deep Learning has been widely used for tackling challenging natural language processing tasks over the recent years. Similarly, the application of Deep Neural Networks in legal analytics has increased significantly. In this survey, we study the early adaptation of Deep Learning in legal analytics focusing on three main fields; text classification, information extraction, and information retrieval. We focus on the semantic feature representations, a key instrument for the successful application of deep learning in natural language processing. (...) Additionally, we share pre-trained legal word embeddings using the word2vec model over large corpora, comprised legislations from UK, EU, Canada, Australia, USA, and Japan among others. (shrink)

The contract review process can be a costly and time-consuming task for lawyers and clients alike, requiring significant effort to identify and evaluate the legal implications of individual clauses. To address this challenge, we propose the use of natural language processing techniques, specifically text classification based on deontic tags, to streamline the process. Our research question is whether natural language processing techniques, specifically dense vector embeddings, can help semi-automate the contract review process and reduce time and costs for (...) legal professionals reviewing deontic modalities in contracts. In this study, we create a domain-specific dataset and train both baseline and neural network models for contract sentence classification. This approach offers a more efficient and cost-effective solution for contract review, mimicking the work of a lawyer. Our approach achieves an accuracy of 0.90, showcasing its effectiveness in identifying and evaluating individual contract sentences. (shrink)

A method to identify ontology components is presented in this article. The method relies on Natural Language Processing (NLP) techniques to extract concepts and relations among these concepts. This method is applied in the legal field to build an ontology dedicated to information retrieval. Legal texts on which the method is performed are carefully chosen as describing and conceptualizing the legal domain. We suggest that this method can help legal ontology designers and may be used (...) while building ontologies dedicated to other tasks than information retrieval. (shrink)

Legal Judgment Prediction (LJP) is a typical application of Artificial Intelligence in the intelligent judiciary. Current research primarily focuses on automatically predicting law articles, charges, and terms of penalty based on the fact description of cases. However, existing methods for LJP have limitations, such as neglecting document structure and ignoring case similarities. We propose a novel framework called Graph Contrastive Learning with Augmentation (GCLA) for legal judgment prediction to address these issues. GCLA constructs trainable document-level graphs for fact (...) description, capturing local and global context through sentence-level subgraphs. Graph augmentation enhances robustness. We introduce a comparison case relation perspective, using graph contrastive learning to model case-text label relationships effectively. Experimental results on real-world datasets demonstrate the competitive performance of GCLA. (shrink)

The number of available legal documents has presented an enormous growth in recent years, and the digital processing of such materials is prompting the necessity of systems that support the automatic relevant information extraction. This work presents a system for ontology-based information extraction from natural language texts, able to identify a set of legal events. The system is based on an innovative methodology based on domain ontology of legal events and a set of linguistic rules, integrated through (...) inference mechanism, resulting in a flexible approach and scalable approach. A case study with the use of documents from the Superior Court in Brazil is related, with satisfactory results in precision and recall. (shrink)

As the labeling costs in legal artificial intelligence tasks are expensive. Therefore, it becomes a challenge to utilize low cost to train a robust model. In this paper, we propose a LAIAugment approach, which aims to enhance the few-shot learning capability in legal artificial intelligence tasks. Specifically, we first use the self-training approach to label the amount of unlabelled data to enhance the feature learning capability of the model. Moreover, we also search for datasets that are similar to (...) the training set by improving the text similarity function. We conducted experimental analyses for three legal artificial intelligence tasks, including evidence extraction, legal element extraction, and case multi-label prediction, which composed of 3500 judgement documents. The experimental results show that the proposed LAIAugment method has an average F1-score of 72.3% on the three legal AI tasks, which is 1.93% higher than the baseline model. At the same time, it shows a huge improvement in few-shot learning. (shrink)

Deep Learning has been widely used for tackling challenging natural language processing tasks over the recent years. Similarly, the application of Deep Neural Networks in legal analytics has increased significantly. In this survey, we study the early adaptation of Deep Learning in legal analytics focusing on three main fields; text classification, information extraction, and information retrieval. We focus on the semantic feature representations, a key instrument for the successful application of deep learning in natural language processing. (...) Additionally, we share pre-trained legal word embeddings using the word2vec model over large corpora, comprised legislations from UK, EU, Canada, Australia, USA, and Japan among others. (shrink)

Automated classification of legal documents has been the subject of extensive research in recent years. However, this is still a challenging task for long documents, since it is difficult for a model to identify the most relevant information for classification. In this paper, we propose a two-stage supervised learning approach for the classification of petitions, a type of legal document that requests a court order. The proposed approach is based on a word-level encoder–decoder Seq2Seq deep neural network, such (...) as a Bidirectional Long Short-Term Memory (BiLSTM) or a Bidirectional Encoder Representations from Transformers (BERT) model, and a document-level Support Vector Machine classifier. To address the challenges posed by the lengthy legal documents, the approach introduces a human-in-the-loop approach, whose task is to localize and tag relevant segments of text in the word-level training part, which dramatically reduces the dimension of the document classifier input vector. We performed experiments to validate our approach using a real-world dataset comprised of 270 intermediate petitions, which were carefully annotated by specialists from the 15th civil unit of the State of Alagoas, Brazil. Our results revealed that both BiLSTM and BERT-Convolutional Neural Networks variants achieved an accuracy of up to 95.49%, and also outperformed baseline classifiers based on the Term Frequency–Inverse Document Frequency test vectorizer. The proposed approach is currently being utilized to automate the aforementioned justice unit, thereby increasing its efficiency in handling repetitive tasks. (shrink)

The common law system is a legal system that values precedent, or previous court decisions, in the resolution of current cases. As the availability of legal documents in digital form has increased, it has become more difficult for legal professionals to manually identify relevant past cases due to the vast amount of data. Researchers have developed automated systems for determining the similarity between legal documents to address this issue. Our research explores various representations of a (...) class='Hi'>legal document and discusses a novel paragraph filtering process to identify key paragraphs using legal citation information to remove unnecessary text paragraphs without disturbing the concept of the legal document. State-of-the-art techniques like TF-IDF, BERT, Legal Bert, Doc2Vec, and Legal-longformer are used for the performance analysis of the proposed approach with document comparison. It has been shown that a model trained on the proposed filtered paragraphs can achieve better results than a model trained on the complete text and can also shorten the document by over 40%. The proposed filtering strategy could be helpful for models like BERT, where the maximum token length is fixed. (shrink)

Socialist courts are supposed to apply the law, not make it, and socialist legality denies judicial decisions any precedential status. In 2011, the Chinese Supreme People’s Court designated selected decisions as Guiding Cases to be referred to by all judges when adjudicating similar disputes. One decade on, the paucity of citations to Guiding Cases has been taken as demonstrating the incongruity of case-based adjudication and the socialist legal tradition. Citations are, however, an imperfect measure of influence. Reproduction of language (...) uniquely traceable to Guiding Cases can also be evidence of their impact on judicial decision-making. We employ a local alignment tool to detect unattributed text reuse of Guiding Cases in local court decisions. Our findings suggest that Guiding Cases are more consequential than commonly assumed, thereby complicating prevailing narratives about the antagonism of socialist legality to case law. (shrink)

I consider the potential of eight text-scaling methods for the analysis of jurisprudential change. I use a small corpus of well-documented German Federal Constitutional Court opinions on European integration to compare the machine-generated scores to scholarly accounts of the case law and legal expert ratings. Naive Bayes, Word2Vec, Correspondence Analysis and Latent Semantic Analysis appear to perform well. Less convincing are the performance of Wordscores, ML Affinity and lexicon-based sentiment analysis. While both the high-dimensionality of judicial texts and (...) the validation of computer-based jurisprudential estimates pose major methodological challenges, I conclude that automated text-scaling methods hold out great promise for legal research. (shrink)

Freedom of information laws promote transparency by permitting individuals and organizations to obtain government documents. However, exemptions from disclosure are necessary to protect privacy and to permit government officials to deliberate freely. Deliberative language is often the most challenging and burdensome exemption to detect, leading to high processing costs and delays in responding to open-records requests. This paper describes a novel deliberative-language detection model trained on a new annotated training set. The deliberative-language detection model is a component of a decision-support (...) system for open-records requests under the US Freedom of Information Act, the FOIA Assistant, that ingests documents responsive to an open-records requests, suggests passages likely to be subject to deliberative language, privacy, or other exemptions, and assists analysts in rapidly redacting suggested passages. The tool’s interface is based on extensive human-factors and usability studies with analysts and is currently in operational testing by multiple US federal agencies. (shrink)

Modeling law search and retrieval as prediction problems has recently emerged as a predominant approach in law intelligence. Focusing on the law article retrieval task, we present a deep learning framework named LamBERTa, which is designed for civil-law codes, and specifically trained on the Italian civil code. To our knowledge, this is the first study proposing an advanced approach to law article prediction for the Italian legal system based on a BERT (Bidirectional Encoder Representations from Transformers) learning (...) framework, which has recently attracted increased attention among deep learning approaches, showing outstanding effectiveness in several natural language processing and learning tasks. We define LamBERTa models by fine-tuning an Italian pre-trained BERT on the Italian civil code or its portions, for law article retrieval as a classification task. One key aspect of our LamBERTa framework is that we conceived it to address an extreme classification scenario, which is characterized by a high number of classes, the few-shot learning problem, and the lack of test query benchmarks for Italian legal prediction tasks. To solve such issues, we define different methods for the unsupervised labeling of the law articles, which can in principle be applied to any law article code system. We provide insights into the explainability and interpretability of our LamBERTa models, and we present an extensive experimental analysis over query sets of different type, for single-label as well as multi-label evaluation tasks. Empirical evidence has shown the effectiveness of LamBERTa, and also its superiority against widely used deep-learning text classifiers and a few-shot learner conceived for an attribute-aware prediction task. (shrink)

Court rulings are among the most important documents in all legal systems. This article describes a study in which natural language processing is used for the automatic characterization of Spanish judgments that deal with the physical custody (joint or individual) of minors. The model was trained to identify a set of elements: the type of custody requested by the plaintiff, the type of custody decided on by the court, and eight of the most commonly used arguments in this type (...) of judgment. Two jurists independently annotated more than 3000 judgments, which were used to train a model based on transformers. The main difficulties encountered in this task were the complexity of the judicial language and the need to work with appellate court rulings that have a more complicated structure than decisions at first instance. For the complete court rulings, the F1 score of the inter-annotator agreement ranged from 0.60 to 0.86 and the Kappa index from 0.33 to 0.73. The F1 score of the agreement between the model and the annotators ranged from 0.66 to 0.93 and the Kappa index from 0.57 to 0.80. These results in which the model performance exceeds even the inter-annotator agreement show the high ability of transformers to identify abstract entities in legal texts. (shrink)

Appellate Court Modifications Extraction consists of, given an Appellate Court decision, identifying the proposed modifications by the upper Court of the lower Court judge’s decision. In this work, we propose a system to extract Appellate Court Modifications for Portuguese. Information extraction for legal texts has been previously addressed using different techniques and for several languages. Our proposal differs from previous work in two ways: our corpus is composed of Brazilian Appellate Court decisions, in which we look for a set (...) of modifications provided by the Court; and to automatically extract the modifications, we use a traditional Machine Learning approach and a Deep Learning approach, both as alternative solutions and as a combined solution. We tackle the Appellate Court Modifications Extraction task, experimenting with a wide variety of methods. In order to train and evaluate the system, we have built the KauaneJunior corpus, using public data disclosed by the Appellate State Court of Rio de Janeiro jurisprudence database. Our best method, which is a Bidirectional Long Short-Term Memory network combined with Conditional Random Fields, obtained an \ score of 94.79%. (shrink)

An important challenge when creating automatically processable laws concerns open-textured terms. The ability to measure open-texture can assist in determining the feasibility of encoding regulation and where additional legal information is required to properly assess a legal issue or dispute. In this article, we propose a novel conceptualisation of open-texture with the aim of determining the extent of open-textured terms in legal documents. We conceptualise open-texture as a lever whose state is impacted by three types of forces: (...) internal forces (the words within the text themselves), external forces (the resources brought to challenge the definition of words), and lateral forces (the merit of such challenges). We tested part of this conceptualisation with 26 participants by investigating agreement in paired annotators. Five key findings emerged. First, agreement on which words are open-texture within a legal text is possible and statistically significant. Second, agreement is even high at an average inter-rater reliability of 0.7 (Cohen’s kappa). Third, when there is agreement on the words, agreement on the Open-Texture Value is high. Fourth, there is a dependence between the Open-Texture Value and reasons invoked behind open-texture. Fifth, involving only four annotators can yield similar results compared to involving twenty more when it comes to only flagging clauses containing open-texture. We conclude the article by discussing limitations of our experiment and which remaining questions in real life cases are still outstanding. (shrink)

Building on the computer science concept of _code smells_, we initiate the study of _law smells_, i.e., patterns in legal texts that pose threats to the comprehensibility and maintainability of the law. With five intuitive law smells as running examples—namely, duplicated phrase, long element, large reference tree, ambiguous syntax, and natural language obsession—, we develop a comprehensive law smell taxonomy. This taxonomy classifies law smells by when they can be detected, which aspects of law they relate to, and how (...) they can be discovered. We introduce text-based and graph-based methods to identify instances of law smells, confirming their utility in practice using the United States Code as a test case. Our work demonstrates how ideas from software engineering can be leveraged to assess and improve the quality of _legal_ code, thus drawing attention to an understudied area in the intersection of law and computer science and highlighting the potential of computational legal drafting. (shrink)

Recently, the pandemic caused by COVID-19 is severe in the entire world. The prevention and control of crimes associated with COVID-19 are critical for controlling the pandemic. Therefore, to provide efficient and convenient intelligent legal knowledge services during the pandemic, we develop an intelligent system for legal information retrieval on the WeChat platform in this paper. The data source we used for training our system is “The typical cases of national procuratorial authorities handling crimes against the prevention (...) and control of the new coronary pneumonia pandemic following the law”, which is published online by the Supreme People’s Procuratorate of the People’s Republic of China. We base our system on convolutional neural network and use the semantic matching mechanism to capture inter-sentence relationship information and make a prediction. Moreover, we introduce an auxiliary learning process to help the network better distinguish the relation between two sentences. Finally, the system uses the trained model to identify the information entered by a user and responds to the user with a reference case similar to the query case and gives the reference legal gist applicable to the query case. (shrink)

We present our method for tackling a legal case retrieval task by introducing our method of encoding documents by summarizing them into continuous vector space via our phrase scoring framework utilizing deep neural networks. On the other hand, we explore the benefits from combining lexical features and latent features generated with neural networks. Our experiments show that lexical features and latent features generated with neural networks complement each other to improve the retrieval system performance. Furthermore, our experimental (...) results suggest the importance of case summarization in different aspects: using provided summaries and performing encoded summarization. Our approach achieved F1 of 65.6% and 57.6% on the experimental datasets of legal case retrieval tasks. (shrink)

To be presented at an International Conference on Law, Language, and Interpretation, at the University of Akureyri, Akureyri, Iceland, April 1-2, 2007.