Weakly-Supervised Relation Extraction in Legal Knowledge Bases

This paper presents a four layer model for working with legal knowledge in expert systems. It distinguishes five sources of knowledge. Four contain basic legal knowledge found in published and unpublished sources. The fifth consists of legal metaknowledge. In the model the four basic legal knowledge sources are placed at the lowest level. The metaknowledge is placed at levels above the other four knowledge sources. The assumption is that the knowledge is represented only once. The use of metaknowledge at various levels should make it possible to use the appropriate knowledge for the problem presented to the system. The knowledge has to be represented as closely to the original format as possible for this purpose. Suitable representation formalisms for the various types of knowledge in the five knowledge sources are discussed. It is not possible to indicate a `best' representation formalism for each knowledge source.

Most of the biomedical knowledge the research community has acquired during the past few decades has been deposited in scientific literature as unstructured text. Converting the unstructured text into the structured form will enable novel methodologies and applications for scientific discovery that can fully harness the power of the existing knowledge. To this end, two fundamental questions need to be addressed: named entity recognition (NER) and relation extraction (RE). NER deals with identifying the concepts or entities in texts, such as diseases, genes/proteins, chemical compounds, etc. while RE aims to extract the relations among these entities. Together, the extracted information forms a knowledge graph (KG) where the nodes are entities in the texts and the edges represent their relationships. KGs can link concepts within existing research to allow researchers to find connections that may have been difficult to discover without them. The LitCoin Natural Language Processing (NLP) Challenge was recently organized by NCATS of NIH and NASA to spur innovation by rewarding the most creative and high-impact uses of biomedical text to create KGs. Our team participated in the challenge and ranked first place. We have applied the methods we developed for the LitCoin NLP challenge to all PubMed abstracts and constructed the largest-scale biomedical KG to date. We show that powerful and versatile query functions can be implemented on top of the KG to enable highly specific and accurate knowledge retrieval and inference of causal and indirect relationships. Citation Format: Xin Sui, Yuan Zhang, Feng Pan, Donghu Sun, Menghan Chung, Jinfeng Zhang. Constructing the largest-scale knowledge graph using all PubMed abstracts and its application for highly specific and accurate knowledge retrieval. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5421.

Objective To improve the handover quality of instruments between operating room and central sterile supply department (CSSD) in special period. Methods The instrument recovery records between operating room and CSSD during special period, including weekends, holidays and night shifts from September to December 2016 and from September to December 2017 in Union Hospital of Tongji Medical College, Huazhong University of Science and Technology were collected. The records from September to December 2016 were assigned as control group, using artificial transition method. The records from September to December 2017 were assigned as intervention group, using using new intelligent information system. The quality of instrument handover and the incidence of adverse events were compared between the two groups. Results After using intelligent information system, the accuracy rate of records and the content of instrument package of the intervention group were both higher than those of the control group, and the differences were statistically significant (P<0.01) . The incidences of instrument loss, instrument defect, occupational exposure and problem traceability in the intervention group were lower than those in the control group, and the differences were statistically significant (P<0.01) . Conclusions The intelligent information system has good application effect on instrument handover between operating room and CSSD in special period. It can solve handover problem in time and can be popularized. Key words: Central supply, hospital; Intelligent information system; Surgical instrument recovery; Special time

Introduction. In recent decades, scholars have shown greater research interest in aspects of legal literacy of young people regarding democratic processes in the Russian society. However, despite increasing relevance of the problem of developing and strengthening the legal awareness of modern young people, including university students, this issue has not been sufficiently studied. Meanwhile, subjective misinterpretation of the concepts of law and justice, a low level of legal culture and legal nihilism can be the causes of youth offences, as well as the factors, which distort the worldview under formation.The present publication is aimed to investigate the role of ideas about the law and justice in the structure of legal awareness of students of higher education institutions.Methodology and research methods. In the course of the empirical research, the following psychodiagnostic methods were employed: the questionnaire created on the basis of L. A. Yasyukova’s and the procedures of free associations. To process the obtained data, the content analysis and the methods of mathematic statistics, the criterion φ* (Fisher’s angular transformation) and the Kolmogorov-Smirnov criterion λ were used.Results and scientific novelty. Theoretical approaches to the problem under discussion have been analysed, according to which legal awareness is determined by the conditions of society. Instability of these conditions generates a “paradoxical personality” (M. I. Yenikeev), who accepts and rejects social norms. It is noted that the Russian legal culture is negatively affected by such factors as the dissonance between legal ideology and legal psychology, weakness of civil society institutions, deep financial and economic crisis, etc.The conducted survey manifested that the specifics of vocational education programme do not have a significant impact on the development of student perceptions of law and justice. The peculiarities of this component of legal awareness of students are revealed. The participants of the survey were convinced that the law should be fair; otherwise, it is not necessary to obey the law. The survey revealed that recognising the priority of legislation over conscience in administrative matters, the respondents believed that only just professionals should be at the head of state, regardless of their moral qualities. According to the respondents, orientation on honor and justice should be the basis of order in the society, but the participants noted the ambiguity of these concepts. On the one hand, students understand justice from the position of social balance (everyone receives what they deserve), on the other hand, they are convinced that social justice is ensured by equality. Young people are interested in the dynamics of the current legislation and believe that everyone should have basic legal knowledge, and ignorance of the law does not exempt a person from this responsibility. The respondents are sure that the law is the same for everyone and is enforceable equally by everyone; moreover, the law is designed to protect the interests of a particular person, not society as a whole. The majority of respondents generated the estimation that the ability to defend own rights is more significant than the performance of duties. The authors of the present research draw a generalised conclusion that university students’ perceptions of law and justice are inconsistent and contradictory.Practical significance. The research results can be used by teachers and psychologists of higher educational institutions, as well as by various experts to develop the programmes for formation of youth legal awareness.

Ignorantia juris neminem excusat: Ignorance of the law does not excuse. We are all presumed to know the law in this country and cannot plead ignorance of it as a defence in a legal dispute. This presumption applies to every citizen. Every professional therefore requires at least a basic legal knowledge; as a matter of sound common-sense we need to know when our actions or omissions have legal consequences. The legal accountability of a professional can be seen, arguably, as the prime form of accountability for professionals.

BackgroundWith the rapid expansion of biomedical literature, biomedical information extraction has attracted increasing attention from researchers. In particular, relation extraction between 2 entities is a long-term research topic.ObjectiveThis study aimed to perform 2 multiclass relation extraction tasks of Biomedical Natural Language Processing Workshop 2019 Open Shared Tasks: relation extraction of Bacteria-Biotope (BB-rel) task and binary relation extraction of plant seed development (SeeDev-binary) task. In essence, these 2 tasks are aimed at extracting the relation between annotated entity pairs from biomedical texts, which is a challenging problem.MethodsTraditional research methods adopted feature- or kernel-based methods and achieved good performance. For these tasks, we propose a deep learning model based on a combination of several distributed features, such as domain-specific word embedding, part-of-speech embedding, entity-type embedding, distance embedding, and position embedding. The multi-head attention mechanism is used to extract the global semantic features of an entire sentence. Meanwhile, we introduced a dependency-type feature and the shortest dependency path connecting 2 candidate entities in the syntactic dependency graph to enrich the feature representation.ResultsExperiments show that our proposed model has excellent performance in biomedical relation extraction, achieving F1 scores of 65.56% and 38.04% on the test sets of the BB-rel and SeeDev-binary tasks. Especially in the SeeDev-binary task, the F1 score of our model is superior to that of other existing models and achieves state-of-the-art performance.ConclusionsWe demonstrated that the multi-head attention mechanism can learn relevant syntactic and semantic features in different representation subspaces and different positions to extract comprehensive feature representation. Moreover, syntactic dependency features can improve the performance of the model by learning dependency relation between the entities in biomedical texts.

Relations extraction is a widely researched topic in nature language processing. However, most of the work in the literature concentrate on the methods that are dealing with single relation between two named entities. In the task of multiple relations extraction, traditional statistic-based methods have difficulties in selecting features and improving the performance of extraction model. In this paper, we presented formal definitions of multiple entities and multiple relations and put forward three labeling methods which were used to label entity categories, relation categories and relation conditions. We also proposed a novel relation extraction model which is based on dynamic long short-term memory network. To train our model, entity feature, entity position feature and part of speech feature are used together. These features are used to describe complex relations and improve the performance of relation extraction model. In the experiments, we classified the corpus into three sets which are composed of 0–20 words, 20–35 words and 35+ words sentences. On conll04.corp, the final precision, recall rate and F-measure reached 72.9, 70.8 and 67.9% respectively.

Current research on knowledge graph construction is focused chiefly on general-purpose fields, whereas constructing knowledge graphs in vertically segmented professional fields faces numerous difficulties. To solve the problems of complex relation types of domain entities, the lack of a large amount of annotated corpus, and the difficulty of extraction, this study proposed a method for constructing domain-annotated datasets based on publicly available texts on the web, which integrates remote supervision and semi-supervision. For the relational triad extraction of a given core entity (an entity lexicon defined semi-automatically by experts), an inflated gate attention network structure for increasing the perceptual field of the model is proposed. In addition, a relational extraction model, Ro-DGANet, was designed based on this structure, incorporating the idea of a probability graph. The Ro-DGANet model was experimentally evaluated on the publicly available Chinese datasets LIC2019 and CHIP2020 and compared with the mainstream relation extraction models, achieving the best results with F1 values of 82.99% and 66.39%, respectively. Finally, the Ro-DGANet model was applied to the relation extraction task of equipment components in industrial scenarios and to the relation extraction task of core knowledge points of programming languages. The analysis results show that the proposed method is applicable to open relation extraction among core entities in different domains with reliable performance and portability.

Предметом вивчення в статті є процеси побудови пояснень щодо запропонованих інтелектуальною інформаційною системою рішень. Метою є розробка методу узгодження пояснень в інтелектуальній інформаційній системі з урахуванням обмежень, які визначаються базовими знаннями щодо об'єктів та процесів у предметної області. Завдання: структуризація процесу побудови пояснень з урахуванням обмежень, представлених знаннями щодо предметної області; визначення аспектів узгодження пояснення; розробка методу узгодження знань щодо пояснення та предметної області. Використовуваними підходами є: підходи до побудови пояснень, підходи до узгодження знань. Отримані наступні результати. Виконано структуризацію процесу побудови пояснень з урахуванням етапу узгодження знань. Виділено аспекти узгодження знань пояснення із вхідними даними в сенсі використання даних для тлумачення, із отриманим інтелектуальною інформаційною системою рішенням в сенсі узгодження з задачами користувача, а також із знаннями предметної області в сенсі обмежень для використання пояснення. Запропоновано метод узгодження пояснення із знаннями щодо предметної області. Висновки. Наукова новизна отриманих результатів полягає в наступному. Запропоновано метод узгодження пояснення із базовими знаннями щодо об'єктів та процесів у предметної області. Метод передбачає ітеративне виконання послідовності узгодження даних щодо процесу прийняття рішень в інтелектуальній системі із знаннями, що описують предметну область, перевірку узгодженості розробленого пояснення із множиною знань щодо предметної області, а також узгодження пояснення з результуючим рішенням інтелектуальної інформаційної системи. В практичному аспекті метод орієнтований на формування підмножини пояснень, що не протирічать базовим з знанням щодо предметної області. Пояснення у складі цієї множини у подальшому можуть бути упорядковані за критерієм ефективності, з урахуванням особливостей задач, для вирішення яких призначені сформовані інтелектуальною системою рішення

The relevance of the program is to implement the client part with a chatbot, which makes the interface simple for the client and helps the administrator to obtain detailed statistics of client actions for further analysis. The intelligent e-commerce information system of the clothing brand uses a database and filters information that meets the user's request. The task of the project is to develop a software product that implements an algorithm that will allow you to select products and add them to the shopping cart. The software product consists of a chatbot and a web application. The chatbot is designed for customers and helps to view available products, select the required number of products and add these products to the cart, also provides an opportunity to view a list of all products in the cart. The web application is designed for the administrator, and allows you to edit content, track customer activity and view the shopping cart. During the completion of this thesis, development and research were conducted, as well as the implementation of a chatbot and a web application that provides information support in e-commerce of the clothing brand. The theoretical concepts of intelligent information system and e-commerce in general are studied and certain aspects that are essential for the topic of qualification work are described in more detail. Information on the basic parameters of the intelligent information system of e-commerce of the clothing brand is given, the analysis of classification and features of analogues is carried out. A comparative analysis of all existing software options to solve the problem and selected the most effective. The implementation of the information system used a functional approach in combination with object-oriented. JavaScript programming language and a number of closely related frameworks have been chosen for development. As a result of the last section, an intelligent information system for e-commerce of the clothing brand was created. The designed structure of the program, database was analyzed and the server part of the system was created. The main functions of the future system, the sequence of processes and the relationships between them are identified. Also analyzed and selected system and software tools needed to implement the tasks. A ready-to-use software product was developed.

Purpose—The purpose of this research is to explore the direct relationship between the emotional intelligence of top management and the quality of strategic decisions they take for their companies. This relationship is further examined by the mediating role of open innovation in the context of intelligent information systems that can impact the way top managers take decisions. This research adopted a survey design as cross-sectional data were taken through questionnaires from top management of the UAE national banks. A final sample size of 213 questionnaires completed by managers was obtained and analyzed. As predicted, there was a strong, positive relationship between managers’ emotional intelligence and the quality of their strategic decisions. Open innovation has revolutionized the way top managers of banks take decisions that are later transformed into policies. Decision-makers are required to possess the skill of decision-making by being vigilant of their surroundings. Hence, they have emotional intelligence and intelligent information systems (IIS) only enhances the trait. IIS is the glorified version of open innovation that further contributes to the decision-making process and the quality of decisions. This research is one of a kind as no one has explored these dimensions of emotional intelligence in the UAE.

The task of entity relation extraction discovers new relation facts and enables broader applications of knowledge graph. Distant supervision is widely adopted for relation extraction, which requires large amounts of texts containing entity pairs as training data. However, in some specific domains such as medical-related applications, entity pairs that have certain relations might not appear together, thus it is difficult to meet the requirement for distantly supervised relation extraction. In the light of this challenge, we propose a novel path-based model to discover new entity relation facts. Instead of finding texts for relation extraction, the proposed method extracts path-only information for entity pairs from the current knowledge graph. For each pair of entities, multiple paths can be extracted, and some of them are more useful for relation extraction than others. In order to capture this observation, we employ attention mechanism to assign different weights for different paths, which highlights the useful paths for entity relation extraction. To demonstrate the effectiveness of the proposed method, we conduct various experiments on a large-scale medical knowledge graph. Compared with the state-of-the-art relation extraction methods using the structure of knowledge graph, the proposed method significantly improves the accuracy of extracted relation facts and achieves the best performance.

Integrating the extracted facts with an existing knowledge base has raised an urgent need to address the problem of entity linking. Specifically, entity linking is the task to link the entity mention in text with the corresponding real world entity in the existing knowledge base. However, this task is challenging due to name ambiguity, textual inconsistency, and lack of world knowledge in the knowledge base. Several methods have been proposed to tackle this problem, but they are largely based on the co-occurrence statistics of terms between the text around the entity mention and the document associated with the entity. In this paper, we propose LINDEN, a novel framework to link named entities in text with a knowledge base unifying Wikipedia and WordNet, by leveraging the rich semantic knowledge embedded in the Wikipedia and the taxonomy of the knowledge base. We extensively evaluate the performance of our proposed LINDEN over two public data sets and empirical results show that LINDEN significantly outperforms the state-of-the-art methods in terms of accuracy.

The BioRED track at BioCreative VIII calls for a community effort to identify, semantically categorize, and highlight the novelty factor of the relationships between biomedical entities in unstructured text. Relation extraction is crucial for many biomedical natural language processing (NLP) applications, from drug discovery to custom medical solutions. The BioRED track simulates a real-world application of biomedical relationship extraction, and as such, considers multiple biomedical entity types, normalized to their specific corresponding database identifiers, as well as defines relationships between them in the documents. The challenge consisted of two subtasks: (i) in Subtask 1, participants were given the article text and human expert annotated entities, and were asked to extract the relation pairs, identify their semantic type and the novelty factor, and (ii) in Subtask 2, participants were given only the article text, and were asked to build an end-to-end system that could identify and categorize the relationships and their novelty. We received a total of 94 submissions from 14 teams worldwide. The highest F-score performances achieved for the Subtask 1 were: 77.17% for relation pair identification, 58.95% for relation type identification, 59.22% for novelty identification, and 44.55% when evaluating all of the above aspects of the comprehensive relation extraction. The highest F-score performances achieved for the Subtask 2 were: 55.84% for relation pair, 43.03% for relation type, 42.74% for novelty, and 32.75% for comprehensive relation extraction. The entire BioRED track dataset and other challenge materials are available at https://ftp.ncbi.nlm.nih.gov/pub/lu/BC8-BioRED-track/ and https://codalab.lisn.upsaclay.fr/competitions/13377 and https://codalab.lisn.upsaclay.fr/competitions/13378. Database URL: https://ftp.ncbi.nlm.nih.gov/pub/lu/BC8-BioRED-track/https://codalab.lisn.upsaclay.fr/competitions/13377https://codalab.lisn.upsaclay.fr/competitions/13378.

Relation extraction is a core task in natural language processing, aiming to identify semantic relationships between entities in unstructured text. Existing relation extraction methods face significant challenges when dealing with overlapping triples, especially in Entity Pair Overlap (EPO) and Single Entity Overlap (SEO) scenarios. This paper proposes a Bi-directional Dynamic Gating Cascaded (B-DGC) relation extraction model, which is improved based on the CasRel model. The model first uses a BERT encoder to obtain text embeddings, then employs a fully connected neural network to simultaneously identify head and tail entities. It fuses text embeddings with head entity features from the head entity sequence to identify corresponding tail entities under specific relationships. Finally, a dynamic gating mechanism is used to fuse the tail entity recognition results from two stages. Experimental results on NYT and WebNLG datasets show that the B-DGC model significantly outperforms the baseline model CasRel in precision, recall, and F1-score, achieving 90.4%, 91.1%, and 90.7% on NYT, and 92.6%, 92.2%, and 92.4% on WebNLG. Additionally, the B-DGC model demonstrates superior performance across various overlapping triple types, confirming the effectiveness of the bi-directional verification and dynamic gating mechanism.