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)

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)

The exponential digitization of medical data has led to a transformation of the practice of medicine. This change notably raises a new complexity of issues surrounding health IT. The proper use of these communication tools, such as telemedicine, e-health, m-health the big medical data, should improve the quality of monitoring and care of patients for an information system to "human face". Faced with these challenges, the author analyses in an ethical angle the patient-physician relationship, sharing, transmission and storage (...) of medical information, setting pins to an ethic for the digitization of medical information. Drawing on good practice recommendations closely associated with values, this model is developing tools for reflection and present the keys to understanding the decision-making issues that reflect both the technological constraints and the complex nature of human reality in medicine. (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 possibility now exists of capturing a cradle-to-grave record of everything a person says or does. No longer must a personal history be a partial picture. Technology has made it possible to record, process, store, and retrieve all the text, sounds, and images that are required to paint a complete picture of an individual’s life. The efforts of future historians will be directed more to forgetting than to remembering. By default, society will forget nothing. For almost all of human history, (...) remembering has meant the judicious selection and organization of observations about events and people. There used to be an information frontier beyond which the past was a tabula rasa . That information frontier has gone the way of the dodo. The social memory of events in an individual’s life is not only detailed but permanent. Although physical storage is fallible and changes in technology may make some devices effectively unreadable, these limitations are more than made up for by the negligible cost of duplication and distribution in a network. The record of one’s triumphs and tragedies will haunt one forever. Gone is personal privacy since facts buried in the past can be uncovered at any moment. Gone is personal memory since it is easier to rely on the external social memory of cyberspace. In what follows, we explain these observations and trace their consequences. (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)

The taxa that appear in biological classifications are commonly seen as representing information about the traits of their member organisms. This paper examines in what way taxa feature in the storage and retrieval of such information. I will argue that taxa do not actually store much information about the traits of their member organisms. Rather, I want to suggest, taxa should be understood as functioning to localize organisms in the genealogical network of life on Earth. (...) Taxa store information about where organisms are localized in the network, which is important background information when it comes to establishing knowledge about organismal traits, but it is not itself information about these traits. The view of species and higher taxa that is proposed here follows from examining three problems that occur in contemporary biological systematics and are discussed here: the problem of generalization over taxa, the problem of phylogenetic inference, and the problematic nature of the Tree of Life. (shrink)

Transcripts of court judgments are rich repositories of legal knowledge, detailing the intricacies of cases and the rationale behind judicial decisions. Key information extracted from these documents provides a concise overview of a case, crucial for both legal experts and the public. With the advent of large language models (LLMs), automatic information extraction has become increasingly feasible and efficient. This paper presents a comprehensive study on the application of GPT-4, a large language model, for information extraction from (...) judgments of the UK Employment Tribunal (UKET). Through a manual verification process, we meticulously evaluated GPT-4’s performance in extracting critical information from these judgments to ensure the accuracy and relevance of the extracted data. Our research is structured around two primary extraction tasks: the first involves a general extraction of eight key aspects that hold significance for both legal specialists and the general public, namely, the facts of the case, the claims made, references to legal statutes, references to precedents, general case outcome and corresponding labels, detailed order and remedies and, finally, reasons for the decision. The second task is more focused, aimed at analysing three of those extracted features, i.e., facts, claims and outcomes, in order to facilitate the development of a tool capable of predicting the outcome of employment law disputes. Through our analysis, we demonstrate that LLMs like GPT-4 can obtain high accuracy in legal information extraction, highlighting the potential of LLMs in revolutionising the way legal information is processed and utilised, offering significant implications for legal research and practice. (shrink)

Norms are essential in our society: they dictate how individuals should behave and interact within a community. They can be written down in laws or other written sources. Interpretations often differ; this is where formalisations offer a solution. They express an interpretation of a source of norms in a transparent manner. However, creating these interpretations is labour intensive. Natural language processing techniques can support this process. Previous work showed the potential of transformer-based models for Dutch law texts. In this paper, (...) we (1) introduce a dataset of 2335 English sentences annotated with legal semantic roles conform the Flint framework; (2) fine-tune a collection of language models on this dataset, and (3) query two non-fine-tuned generative large language models (LLMs). This allows us to compare performance of fine-tuned domain-specific, task-specific, and general language models with non-fine-tuned generative LLMs. The results show that models fine-tuned on our dataset have the best performance (accuracy around 0.88). Furthermore, domain-specific models perform better than general models, indicating that domain knowledge is of added value for this task. Finally, different methods of querying LLMs perform unsatisfactorily, with maximum accuracy scores around 0.6. This indicates that for specific tasks, such as this adaptation of semantic role labelling, the process of annotating data and fine-tuning a smaller language model is preferred over querying a generative LLM, especially when domain-specific models are available. (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)

In recent years, with the rapid development of computer storage capabilities and network transmission capabilities, users can easily share their own video and image information on social networking sites, and the amount of multimedia data on the network is rapidly increasing. With the continuous increase of the amount of data in the network, the establishment of effective automated data management methods and search methods has become an increasingly urgent need. This paper proposes a retrieval method of human (...) motion data based on motion capture in index space. By extracting key frames from the original motion to perform horizontal dimensionality reduction and defining features based on Laban motion analysis, the motion segment is subjected to vertical feature dimensionality reduction. After extracting features from the input motion segment, motion matching is performed on the index space. This paper designs the optimization method of the phased dynamic time deformation algorithm in time efficiency and analyzes the optimization method of the phased dynamic time deformation algorithm in time complexity. Considering the time efficiency redundancy, this paper optimizes the time complexity of the phased dynamic time deformation method. This improves the time efficiency of the staged dynamic time warping algorithm, making it suitable for larger-scale human motion data problems. Experiments show that the method in this paper has the advantage of speed, is more in line with the semantics of human motion, and can meet the retrieval requirements of human motion databases. (shrink)

The design of Web browsers has resulted in a transfer of power to Web users and developers who often lack an ethical framework in which to act. For example, the technology makes it simple to copy and use other people’s Web page formatting without their permission. The author argues that we need to educate more people about ethical Web practices, and the author asks for “rules of the road” which amateurs and professionals can understand and follow. This article discusses four (...) areas of concern about Web development: the browser wars, information storage and retrieval, access for the handicapped, and cookies. For teachers, there are suggestions on how to use browsers to help students learn about Web ethics. (shrink)

With the advancement of scientific collaboration in the 20th century, researchers started collaborating in many research areas. Researchers and scientists no longer remain solitary individuals; instead, they collaborate to advance fundamental understandings of research topics. Various bibliometric methods are used to quantify the scientific collaboration among researchers and scientific communities. Among these different bibliometric methods, the co-authorship method is one of the most verifiable methods to quantify or analyze scientific collaboration. In this research, the initial study has been conducted to (...) analyze interdisciplinary research activities in the computer science domain. The ACM has classified the computer science fields. We selected the Journal of Universal Computer Science for experimentation purposes. The J.UCS is the first Journal of Computer Science that addresses a complete ACM topic. Using J.UCS data, the co-authorship network of the researcher up to the 2nd level was developed. Then the co-authorship network was analyzed to find interdisciplinary among scientific communities. Additionally, the results are also visualized to comprehend the interdisciplinary among the ACM categories. A whole working web-based system has been developed, and a forced directed graph technique has been implemented to understand IDR trends in ACM categories. Finally, the IDR values between the categories are computed to quantify the collaboration trends among the ACM categories. It was found that “Artificial Intelligence” and “Information Storage and Retrieval”, “Natural Language Processing and Information Storage and Retrieval”, and “Human-Computer Interface” and “Database Applications” were found the most overlapping areas by acquiring an IDR score of 0.879, 0.711, and 0.663, respectively. (shrink)

My aim in this paper is to begin a discussion about how, and to what extent, Martin Heidegger’s thinking about technology offers helpful critical terms for thinking about the nature and global sway of today’s most dominant and prevalent forms of technology, namely the interrelated technologies of information, communication, and (capitalist) commerce. My suggestion will be that Heidegger’s thought does indeed have implications for critical thinking about these technologies, but that in order to see how it does, we may (...) have to deepen and further radicalize some of Heidegger’s suggestions about essence, being, and form. This need for deepening is connected with the way in which these specific technologies themselves depend on the character and structure of language , in that they present themselves as effective means for the manipulation, storage, retrieval and exchange of broadly linguistic, symbolic, and informational material. In fact, in the context of Heidegger’s late thought about technology as well as language, the syntagm “information technology” itself – a term that, as far as I know, Heidegger himself never used – nevertheless points the way, as I shall argue, to a deepening of the Heideggerian inquiry into form which itself can facilitate an improved critical understanding of the implications of these technologies for life around the planet today. (shrink)

A view of evolution is presented in this paper (a two paper series), intended as a methodological infrastructure for modeling spatio-cultural systems (the design outline of such a model is presented in paper II). A motivation for the re-articulation of evolution as information dynamics is the phenomenologically discovered prerequisite of embedding a meaning-attributing apparatus in any and all models of spatio-cultural systems. An evolution is construed as the dynamics of a complex system comprised of memory devices, connected in an (...) ordered fashion (not randomly) by information-exchanges. An information-exchange transpires when the recipient system adopts a strategy (a continuum of events) that eventually changes its structure; namely, after the exchange, it contains and conveys different information. These memory devices—sub-systems—are also similarly constructed complex system. Only a part of the information is retained by a system in its physical-memory storage, which eventually loses this function too, when the ability to retrieve a common enough structure is lost. The entire amount of information is a system’s structure of connections (information exchanges); it is contained (apparently stored) dynamically when a system is observed in a temporarily stable state. This temporary permanence—robustness and resilience—is attained dynamically; namely, enabled by changes taking place in its sub-systems and in each of their sub-systems. Therefore, for modeling such a system a multi-layer/multi-scale approach is preferable. It enables the addition and subtraction of an interim scale and the consideration of such a scale as a micro or a macro (thus initiating a maneuver up or down scale) according to an ad-hoc requirement of the model, which imitates an envisaged sub system (called an ‘Inner World’), to which a certain range of decisions is relegated. This dynamics is driven (in time) by dynamically originated information growth, as defined by Shannon; i.e. by the fact that each of certain state transitions have occurred sometime in the past of the system just so and not otherwise, and by the fact that they have occurred in a certain order. Therefore, each ‘history’ and memory retrieval availability of information is unique, and thus can be used to differentiate meanings. Hence, there cannot be a comprehensive solution to the meaning attribution model-design challenge. However, the observation that at the core of each envisaged complex system, moving in time according to a rounded logic, there is an information manipulating device, operating necessarily according to a Boolean logic, can be copied into the design of a model. This observation enables, therefore, the embedding of a specific, locally fitted, meaning attributing device, which is an information manipulating mechanism (it splices/attaches one segment of information to another—its meaning). However, this is just a framework; the actual solution has still to be found locally—for each subject system. Such a solution is demonstrated for the change in location or layout in the Israeli city in paper II. (shrink)

People typically remember emotionally negative words better than neutral words. Two experiments are reported that investigate whether emotionally enhanced memory for negatively arousing words is based on a storage or retrieval advantage. Participants studied non-word–word pairs that either involved negatively arousing or neutral target words. Memory for these target words was tested by means of a recognition test and a cued-recall test. Data were analysed with a multinomial model that allows the disentanglement of storage and retrieval (...) processes in the present recognition-then-cued-recall paradigm. In both experiments the multinomial analyses revealed no storage differences between negatively arousing and neutral words but a clear retrieval advantage for negatively arousing words in the cued-recall test. These findings suggest that EEM for negatively arousing words is driven by associative processes. (shrink)

This paper adopts Hadoop to build and test the storage and retrieval platform for painting resources. This paper adopts Hadoop as the platform and MapReduce as the computing framework and uses Hadoop Distributed Filesystem distributed file system to store massive log data, which solves the storage problem of massive data. According to the business requirements of the system, this paper designs the system according to the process of web text mining, mainly divided into log data preprocessing module, (...) log data storage module, log data analysis module, and log data visualization module. The core part of the system is the log data analysis module. The analysis of search keywords ranking, Uniform Resource Locator, and user click relationship, URL ranking, and other dimensions are realized through data statistical analysis, and Canopy coarse clustering is performed first according to search keywords, and then K-means clustering is used for the results after Canopy clustering, and the calculation of cosine similarity is adopted to realize the grouping of users and build user portrait. The Hadoop development environment is installed and deployed, and functional and performance tests are conducted on the contents implemented in this system. The constructed private cloud platform for remote sensing image data can realize online retrieval of remote sensing image metadata and fast download of remote sensing image data and solve the problems in storage, data sharing, and management of remote sensing image data to a certain extent. (shrink)

We define and study the notions of stability and relevance for precedent-based reasoning, focusing on Horty’s result model of precedential constraint. According to this model, precedents constrain the possible outcomes for a focus case, which is a yet undecided case, where precedents and the focus case are compared on their characteristics (called dimensions). In this paper, we refer to the enforced outcome for the focus case as its justification status. In contrast to earlier work, we do not assume that all (...) dimension values of the focus case or the precedent cases have been established with certainty: rather, each dimension is assigned a set of possible values. We define a focus case as stable if its justification status is the same for every choice of the possible values. For focus cases that are not stable, we study the task of identifying relevance: which possible values should be excluded to make the focus case stable? In addition, we introduce the notion of possibility to verify if a user can assign an outcome to an unstable focus case without making the case base of precedents inconsistent. We show how the tasks of identifying justification, stability, relevance and possibility can be applied for human-in-the-loop decision support. Finally, we discuss the computational complexity of these tasks and provide efficient algorithms. (shrink)

How much information from experience does a normal adult remember? The “functional information content” of human memory was estimated in several ways. The methods depend on measured rates of input and loss from very long‐ term memory and on analyses of the informational demands of human memory‐based performance. Estimates ranged around 109 bits. It is speculated that the flexible and creative retrieval of facts by humans is a function of a large ratio of “hardware” capacity to functional (...) storage requirements. (shrink)

In lieu of an abstract, here is a brief excerpt of the content:Ideas in the Brain: The Localization of Memory Traces in the Eighteenth CenturyTimo KaitaroPlato suggests in the Theaetetus that we imagine a piece of wax in our soul, a gift from the goddess of Memory. We are able to remember things when our perceptions or thoughts imprint a trace upon this piece of wax, in the same manner as a seal is stamped on wax. Plato uses this metaphor (...) to explain the errors which arise when we mistake something for something else: we connect the perception of an object with the trace belonging to another. The metaphor can also be used in explaining differences in people’s mnestic capacities: rapid learning and forgetting correspond to soft wax, impure wax results in muddled traces, etc.1 If we locate the traces in the brain instead of in the soul, Plato’s metaphor gains consistency and turns into a testable hypothesis. This move was already made by Quintilian.2So, the metaphor of memory as traces in the brain is evidently not a modern invention. This is easy to understand. To frame the hypothesis one needs only to reflect on how we use objects outside our brains for mnemonic purposes. We conserve ideas by tracing or printing letters and words on paper. We are also able to conserve images in drawings, paintings and prints. What could be more natural than to think of memory as the formation of traces in the brain? The development of even better information storage techniques in the form of pictures, symbols, and signs provides the metaphor extra plausibility. Plato’s seal allows us to imagine the imprint of a person’s likeness, but modern techniques of information storage and retrieval make it even easier to imagine memory in general as the formation of material traces.Explaining memory in terms of material traces could, of course, be taken [End Page 301] to mean that the formation of material traces in the brain is merely a necessary condition of memory. But one could also take a more reductionistic stance which identifies memory with the formation of material traces. Furthermore, one could identify specific mental items (sensations, ideas, memories) with specific physiological events or anatomical entities in the brain. This seems especially plausible if we imagine that the traces in the brain are as discrete and separate as ideas or memories are in our minds. Since the traces we form or print on paper consist of discrete letters and words, the thought that the traces in the brain are equally discrete and separate comes easily. Plato, in fact, made separateness a condition for the memory to work properly: it is not easy to read signs printed on one another.3 Of course, no traces are actually visible in the brain. But they are easy to imagine, as it is to imagine explanations of psychological phenomena, such as association, based on these traces.At the end of the seventeenth century, explanations of mental phenomena referring to material traces in the brain were used by writers, from Platonists to materialists, who supported widely different theories about the nature of the human mind.4 In fact, as I will show later in this paper, the question of the possible identity of mental entities, like ideas or sensations, with material entities or traces in the brain is relatively independent of the question of the existence of an immaterial soul. Or rather, if there is a systematic connection between the answers proposed to these two questions in the eighteenth century, it is not the connection one would expect. Nowadays we often tend to think of a prototypical materialist as someone who makes radical reductionistic claims about the identity of mental phenomena with brain states. We also tend to think of the eighteenth-century French materialists as “mechanistic.”5 In view of this reputation, it may come as a surprise to find that some eighteenth-century materialists, notably Denis Diderot, were in fact consistent anti-reductionists and opposed to mechanical or mechanistic explanations of the mental.6 In fact, the epithet “mechanistic” is misleading even when applied to the philosopher who is traditionally presented as the paradigmatic example of a “mechanistic materialist,” Julien Offray... (shrink)

Research on action memory has been pursued for more than 30 years, but it is still unclear what drives the recollection process of performed actions. In this study, we used the remember/know paradigm and designed two experiments to examine the relation between item-specific processing and retrieval awareness of subject-performed tasks (SPT). The results showed that SPT allows remember responses in remember-know judgments more easily; that is, SPT can enhance the frequency of recalling re-collective experience. Item-specific processing can improve the (...) memory performance and the proportion of remember judgments of verbal tasks (VT), but it does not improve the memory performance and proportion of remember judgments of SPT, indicating that SPT can enhance item-specific processing, which leads to more remember responses in judgment. The relation between item-specific processing and retrieval awareness of SPT is also discussed. (shrink)

Episodic memory is the form of memory involved in remembering personally experienced past events. Here, I address two questions about episodic memory’s function: what does episodic memory do for us, and why do we have it? Recent work addressing these questions has emphasized episodic memory’s role in imaginative simulation, criticizing the mnemonic view on which episodic memory is “for” remembering. In this paper, I offer a defense of the mnemonic view by highlighting an underexplored mnemonic function of episodic memory – (...) namely, its role in the encoding, storage and retrieval of the type of information more standardly associated with semantic memory. I argue that in healthy individuals, episodic memory plays a central role in the encoding, storage and retrieval of prototypically semantic information, analogous to the role played by mind palaces in the method of loci, and may have been selected on for this reason. This suggests new directions for studying episodic memory, particularly in nonhuman animals. (shrink)

As quantum information science approaches the goal of constructing quantum computers, understanding loss of information through decoherence becomes increasingly important. The information about a system that can be obtained from its environment can facilitate quantum control and error correction. Moreover, observers gain most of their information indirectly, by monitoring (primarily photon) environments of the “objects of interest.” Exactly how this information is inscribed in the environment is essential for the emergence of “the classical” from the (...) quantum substrate. In this paper, we examine how many-qubit (or many-spin) environments can store information about a single system. The information lost to the environment can be stored redundantly, or it can be encoded in entangled modes of the environment. We go on to show that randomly chosen states of the environment almost always encode the information so that an observer must capture a majority of the environment to deduce the system’s state. Conversely, in the states produced by a typical decoherence process, information about a particular observable of the system is stored redundantly. This selective proliferation of “the fittest information” (known as Quantum Darwinism) plays a key role in choosing the preferred, effectively classical observables of macroscopic systems. The developing appreciation that the environment functions not just as a garbage dump, but as a communication channel, is extending our understanding of the environment’s role in the quantum-classical transition beyond the traditional paradigm of decoherence. (shrink)

The appropriate understanding and fast processing of lengthy legal documents are computationally challenging problems. Designing efficient automatic summarization techniques can potentially be the key to deal with such issues. Extractive summarization is one of the most popular approaches for forming summaries out of such lengthy documents, via the process of summary-relevant sentence selection. An efficient application of this approach involves appropriate scoring of sentences, which helps in the identification of more informative and essential sentences from the document. In this work, (...) a novel sentence scoring approach DCESumm is proposed which consists of supervised sentence-level summary relevance prediction, as well as unsupervised clustering-based document-level score enhancement. Experimental results on two legal document summarization datasets, BillSum and Forum of Information Retrieval Evaluation (FIRE), reveal that the proposed approach can achieve significant improvements over the current state-of-the-art approaches. More specifically it achieves ROUGE metric F1-score improvements of (1−6)% and (6−12)% for the BillSum and FIRE test sets respectively. Such impressive summarization results suggest the usefulness of the proposed approach in finding the gist of a lengthy legal document, thereby providing crucial assistance to legal practitioners. (shrink)

The proliferation of legal documents in various formats and their dispersion across multiple courts present a significant challenge for users seeking precise matches to their information requirements. Despite notable advancements in legal information retrieval systems, research into legal recommender systems remains limited. A plausible factor contributing to this scarcity could be the absence of extensive publicly accessible datasets or benchmarks. While a few studies have emerged in this field, a comprehensive analysis of the distinct attributes of legal (...) data that influence the design of effective legal recommenders is notably absent in the current literature. This paper addresses this gap by initially amassing a comprehensive session-based dataset from Jusbrasil, one of Brazil’s largest online legal platforms. Subsequently, we scrutinize and discourse key facets of legal session-based recommendation data, including session duration, types of recommendable legal artifacts, coverage, and popularity. Furthermore, we introduce the first session-based recommendation benchmark tailored to the legal domain, shedding light on the performance and constraints of several renowned session-based recommendation approaches. These evaluations are based on real-world data sourced from Jusbrasil. (shrink)

Many countries employ systemic methods of protecting consumers from unfair business practices. One such practice is the use of abusive clauses in business-to-consumer (B2C) contracts, which unfairly impose additional obligations on the consumer or deprive them of their due rights. This article presents an information system that utilizes artificial intelligence methods to automate contract analysis and to detect abusive clauses. The goal of the system is to support the entire administrative process, from contract acquisition, through text extraction and the (...) recommendation of potentially abusive clauses, to the generation of official administrative documents that can be sent to court or to the owners of firms. This article focuses on the components that use machine learning methods. The first is an intelligent crawler that is responsible for automatically detecting contract templates on websites and retrieving them into the system. The second is a document analysis module that implements a clause recommendation algorithm. The algorithm employs transformer-based language models and information retrieval methods to identify abusive passages in text. Our solution achieved first place in a competition on the automatic analysis of B2C contracts organized by the Polish Office of Competition and Consumer Protection (UOKiK), and has since been implemented as an official tool to support the contract analysis process in Poland. (shrink)

The never ending growth of digital information and the availability of low-cost storage facilities and networks capacity is leading users towards moving their data to remote storage resources. Since users’ data often holds identity-related information, several privacy issues arise when data can be stored in untrusted domains. In addition digital identity management is becoming extremely complicated due to the identity replicas proliferation necessary to get authentication in different domains. GMail and Amazon Web Services, for instance, are (...) two examples of online services adopted by million of users throughout the world which hold huge amounts of sensitive users data. State-of-the-art encryption tools for large-scale distributed infrastructures allow users to encrypt content locally before storing it on a remote untrusted repository. This approach can experience performance drawbacks, when very large data-sets must be encrypted/decrypted on a single machine. The proposed approach extends the existing solutions by providing two additional features: (1) the encryption can also be delegated to a pool of remote trusted computing resources, and (2) the definition of the encryption context which drives the tool to select the best strategy to process the data. The performance benchmarks are based on the results of tests carried out both on a local workstation and on the Grid INFN Laboratory for Dissemination Activities (GILDA) testbed. (shrink)

Knowledge based privacy policies are more declarative than traditional action based ones, because they specify only what is permitted or forbidden to know, and leave the derivation of the permitted actions to a security monitor. This inference problem is already non trivial with a static privacy policy, and becomes challenging when privacy policies can change over time. We therefore introduce a dynamic modal logic that permits not only to reason about permitted and forbidden knowledge to derive the permitted actions, but (...) also to represent explicitly the declarative privacy policies together with their dynamics. The logic can be used to check both regulatory and behavioral compliance, respectively by checking that the permissions and obligations set up by the security monitor of an organization are not in conflict with the privacy policies, and by checking that these obligations are indeed enforced. (shrink)

The structures of discourse used by legal and ordinary languages share differences that foster technical issues when applying or fine-tuning general-purpose language models for open-domain question answering on legal resources. For example, longer sentences may be preferred in European laws (i.e., Brussels I bis Regulation EU 1215/2012) to reduce potential ambiguities and improve comprehensibility, distracting a language model trained on ordinary English. In this article, we investigate some mechanisms to isolate and capture the discursive patterns of legalese in order to (...) perform zero-shot question answering, i.e., without training on legal documents. Specifically, we use pre-trained open-domain answer retrieval systems and study what happens when changing the type of information to consider for retrieval. Indeed, by selecting only the important parts of discourse (e.g., elementary units of discourse, EDU for short, or abstract representations of meaning, AMR for short), we should be able to help the answer retriever identify the elements of interest. Hence, with this paper, we publish Q4EU, a new evaluation dataset that includes more than 70 questions and 200 answers on 6 different European norms, and study what happens to a baseline system when only EDUs or AMRs are used during information retrieval. Our results show that the versions using EDUs are overall the best, leading to state-of-the-art F1, precision, NDCG and MRR scores. (shrink)

In order to distribute our thoughts and feelings, we must make intelligible and distributable copies of them. From approximately 1375 to 1450, certain Europeans started fully mechanized replication of texts and images, based on predecessor “smaller” technologies. What they started became the most powerful means for the distribution, storage, and retrieval of knowledge in history, up until the invention of digital means. We have scant information about the initiation of print technologies in the period up to Gutenberg, (...) and the picture of Gutenberg that we have has become a great deal more complicated than hitherto. There has not been, however, an approach to the “pre-printing” period in terms of the history of idea or intellectual history. After a brief survey of established approaches, this essay argues that distribution by impression, or print, is bound up with ancient metaphors for understanding communication by the making of multiples. I suggest that there is a rich field of study for printing history in the sophisticated concepts of reality that medieval and late Scholastic philosophy developed. These concepts helped to express and develop a desire or need for communication that led to the technology of replicating texts and images for wide and continued distribution. (shrink)

In lieu of an abstract, here is a brief excerpt of the content:Mild Cognitive Impairment:What's in a Name?Steven R. Sabat (bio)Keywordslabeling, mild cognitive impairment, recall memory, selfhood, stereotype threatCorner and Bond (2006) raise a number of important conceptual issues related to the problems involved in defining mild cognitive impairment (MCI), differentiating it from normal aging, the definition of normal aging itself, and ethical issues surrounding the possible adverse effects of a diagnosis of MCI on the individuals thus described. It would (...) seem that portentous ramifications are involved and that clarifying the conceptual issues would be of great value, not only to researchers but also to the ever-growing population of elderly people. In what follows, I address some of these issues further.Problems of Definition and MeasurementOne of the problems that the authors discuss at the outset of their article concerns what methods are appropriate for detecting those with MCI: "Neuropsychological tests alone do not discriminate between normal aged-related memory loss and MCI or dementia and clinical judgment remains central to diagnosis" (Corner and Bond 2006, 5). Of course, to be able to discriminate between normal aged-related memory problems and MCI, we must first be able to define the former. What exactly is normal? As the authors correctly point out, the term has different meanings, one being normative (usual), another being nonpathologic. Is it normal (usual) for some sort of breakdown to occur in brain systems as a function of increased age? If so, is there a threshold age below which the very same physiologic breakdown is unusual? Would such a threshold age, if one existed, be the same for everyone, regardless of genotype and the cumulative interactive effects of nutrition, health, and stress-related factors over a period of decades? In effect, is some sort of memory impairment part of the natural (usual) process of aging for some people at particular ages rather than for others? And exactly what does it mean to be mildly impaired in the cognitive domain?Hogan and McKeith (2001) presented a working group's definition characterizing MCI such that people who complain about memory problems and who also provide objective evidence of such problems, yet who simultaneously display "normal general cognitive function and intact activities of daily living" (Hogan and McKeith 2001, 1131), should be identified and followed. Here again, however, we confront the problem of defining "normal general cognitive function" as well as "normal activities of daily living." Can one have normal general cognitive function without normal memory function? If so, what exactly constitutes [End Page 13] general cognitive function and on what conceptual or theoretical basis is such a definition valid? Furthermore, what sort of memory function are we talking about? Memory can be said to involve a number of stages: encoding, storage, and retrieval. Retrieval can be of the explicit type (conscious recall or recognition of specific information) with which people with Alzheimer's disease (AD) often reveal deficits or of the implicit type (a change in behavior as a result of previous experience of which the person may have no conscious recollection) as demonstrated in people with AD (Knopman and Nissen 1987). That is, people with AD can display deficits in explicit memory functions without similar deficits simultaneously in implicit memory functions.A partial answer to the question about memory loss in MCI may be found in the use of the Mini-Mental State Examination (MMSE; Folstein, Folstein, and McHugh 1975), originally intended to be used as a screening device, but employed more and more frequently as an outcome measure to assess cognitive decline as well as the efficacy of drug treatments. Although the examination purports to test orientation, registration, attention, calculation, recall, language (naming, repeating, reading, writing, three-stage command), and construction, a task analysis reveals that the test is not quite as "global" in the cognitive domain as it is purported to be. For example, ten of the maximum thirty points on the examination are earned for "orientation" in time and place. Yet, all of the questions in this section of the test require intact recall of the relevant information. For example, if a person is asked, "What day of the week is this?" and replies "I don't know," the... (shrink)