Types Of Entities Research Articles

Inductive Meta-Path Learning for Schema-Complex Heterogeneous Information Networks.

Heterogeneous Information Networks (HINs) are information networks with multiple types of nodes and edges. The concept of meta-path, i.e., a sequence of entity types and relation types connecting two entities, is proposed to provide the meta-level explainable semantics for various HIN tasks. Traditionally, meta-paths are primarily used for schema-simple HINs, e.g., bibliographic networks with only a few entity types, where meta-paths are often enumerated with domain knowledge. However, the adoption of meta-paths for schema-complex HINs, such as knowledge bases (KBs) with hundreds of entity and relation types, has been limited due to the computational complexity associated with meta-path enumeration. Additionally, effectively assessing meta-paths requires enumerating relevant path instances, which adds further complexity to the meta-path learning process. To address these challenges, we propose SchemaWalk, an inductive meta-path learning framework for schema-complex HINs. We represent meta-paths with schema-level representations to support the learning of the scores of meta-paths for varying relations, mitigating the need of exhaustive path instance enumeration for each relation. Further, we design a reinforcement-learning based path-finding agent, which directly navigates the network schema (i.e., schema graph) to learn policies for establishing meta-paths with high coverage and confidence for multiple relations. Extensive experiments on real data sets demonstrate the effectiveness of our proposed paradigm.

Bridging the gap: An environmental scan of post-overdose program features and evaluation measures.

Prior overdose incident is a predictor of future non-fatal and fatal overdoses. Therefore, the period immediately following a non-fatal overdose is critical for engaging with survivors; however, they often do not access or receive services immediately after experiencing an overdose. To efficiently reach this high-risk population and provide survivors with services, novel interventions including post-overdose outreach programs have been developed. Using PRISMA guidelines, we conducted an environmental scan of the academic and gray literature on post-overdose programs to identify key features and measures used to evaluate post-overdose programs. Using the Donabedian quality framework we then analyzed the frequency of measures by type of initiating stakeholder. Of the 36 studies included for final extraction, 14 were descriptive studies, ten were evaluation studies, eight studies identified facilitators and barriers, and four were program design toolkits. Initiator entity types included emergency departments (EDs; n = 10), public health agencies (n = 5), community-based organizations (n = 4), emergency medical services (EMS) (n = 4), law enforcement agencies (n = 4), and some combination or unknown (n = 9). We identified six key features of post-overdose programs across the literature: program theory, team composition, identifying and contacting clients, service provision, public private partnerships, and referrals. We extracted 223 operationalized measures, of which we analyzed 173 and categorized them as client experience (n = 2), structural (n = 12), process (n = 69), and outcome measures (n = 90). Public health agency-initiated programs reported all the experience-related measures (n = 2) and the most outcome measures (n = 23). Programs initiated by EDs reported the most process measures (n = 20), whereas community-based agencies reported the most structural measures (n = 12). Only two of the 223 measures related to stigma. There are six common features across all post-overdose programs and different initiating organization types had differing foci for their evaluation measures. The key features and evaluation measure clusters identified in this environmental scan can help inform future post-overdose programs assess the comprehensiveness of their evaluations, inform funding agencies on the range of existing measures (and gaps) to consider, and support evaluators in study design.

A systematic literature review of recent advances on context-aware recommender systems

AbstractRecommender systems are software mechanisms whose usage is to offer suggestions for different types of entities like products, services, or contacts that could be useful or interesting for a specific user. Other ways have been explored in the field to enhance the power of these systems by integrating the context as an additional attribute. This inclusion tries to extract the user preferences more accurately taking into account multiple components such as temporal, spatial, or social ones. Notwithstanding the magnitude of context-awareness in this area, the research community is in agreement with the lack of framework for context information and how to integrate it into recommender systems. Under this premise, this paper focuses on a comprehensive systematic literature review of the state-of-the-art recommendation techniques and their characteristics to benefit from contextual information. The following survey presents the following contributions as outcomes of our study: (i) determine a framework where multiple aspects are taken into account to have a clear definition of context representation, (ii) the techniques used to incorporate context, and (iii) the evaluation of these methods in terms of reproducibility and effectiveness. Our review also covers some crucial topics about context integration, classification of the contexts, application domains, and evaluation of the used datasets, metrics, and code implementations, where we observed clear shiftings in algorithmic and evaluation trends towards Neural Network approaches and ranking metrics, respectively. Just as importantly, future research opportunities and directions are exposed as final closure, standing out the exploitation of various data sources and the scalability and customization of existing solutions.

An entity and relation extraction model based on context query and axial attention towards patent texts

Patent Entity and Relation Extraction (PERE) aims to extract entities and entity-relation triples from unstructured patent texts. PERE is one of the fundamental tasks in patent text mining, providing crucial technical support for patent retrieval and technology opportunity discovery. Previous works struggle to capture the implicit semantic information hidden within overlapping triples, especially a large number of overlapping triples existing in patent texts. A Patent Entity and Relation Extraction model based on Context query and Axial attention is proposed, named PERE-CA. As for entity recognition, the text segment is regarded as candidate entity span and entity types are acquired by span classification. Subsequently, the semantic context related to an entity pair is calculated by a context query method. And the semantic context is integrated into entity pair representation. For relation extraction, axial attention is implemented to get the implicit semantic information among overlapping entity pairs. And then, the model outputs all valid entity-relation triples. Experimental results on the patent dataset TFH-2020 and the public dataset SciERC demonstrate that the implementation of context query and axial attention can effectively improve extraction performance.

Exploring Green HR Practices in the Construction Industry—Evidence from Poland

The objective of this paper is twofold. First, it systematizes knowledge about green management practices as a manifestation of the green human resource management (HRM) concept. Second, it diagnoses the practices of green management in construction companies in Poland. The research is qualitative and was conducted in three construction companies (one large, one medium, and one small-sized). Due to the small sample size, the study is treated as a pilot study. The research concluded that although green management practices are employed in the studied companies, they are neither systemic nor satisfactory in intensity. In the large company, green management practices are visible at all stages of the human resource process except for employee appraisal. In the medium and small companies, green management is limited to the development of green knowledge and employee awareness activities. The added value of this paper is an original tool for diagnosing the level of advancement of green management practices in competence management, which can be used in various types of business entities, regardless of the sector or industry.

A Chinese named entity recognition method for landslide geological disasters based on deep learning

Landslide Named Entity Recognition (LNER) involves extracting specific entities from Chinese unstructured landslide disaster texts, which is crucial for constructing a knowledge graph and supporting landslide prevention efforts. This study proposes a deep learning-based LNER model that utilizes Bidirectional Encoder Representations from Transformer (BERT) for word embeddings and integrates the Conditional Random Fields (CRF) algorithm and projected gradient descent (PGD) adversarial neural networks to enhance sequence labeling accuracy. The practical implications of this research lie in improving the efficiency and precision of disaster information extraction, aiding in real-time decision-making and risk mitigation strategies. Experiments on the constructed dataset show that the model effectively identifies eight types of landslide entities, achieving a highest F1 score of 89.7%.

Generative named entity recognition framework for Chinese legal domain

Named entity recognition (NER) is a crucial task in natural language processing, particularly challenging in the legal domain due to the intricate and lengthy nature of legal entities. Existing methods often struggle with accurately identifying entity boundaries and types in legal texts. To address these challenges, we propose a novel sequence-to-sequence framework designed specifically for the legal domain. This framework features an entity-type-aware module that leverages contrastive learning to enhance the prediction of entity types. Additionally, we incorporate a decoder with a copy mechanism that accurately identifies complex legal entities without the need for explicit tagging schemas. Our extensive experiments on two legal datasets show that our framework significantly outperforms state-of-the-art methods, achieving notable improvements in precision, recall, and F1 score. This demonstrates the effectiveness of our approach in improving entity recognition in legal texts, offering a promising direction for future research in legal NER.

A Unified Model for Chinese Cyber Threat Intelligence Flat Entity and Nested Entity Recognition

In recent years, as cybersecurity threats have become increasingly severe and cyberattacks have occurred frequently, higher requirements have been put forward for cybersecurity protection. Therefore, the Named Entity Recognition (NER) technique, which is the cornerstone of Cyber Threat Intelligence (CTI) analysis, is particularly important. However, most existing NER studies are limited to recognizing single-layer flat entities, ignoring the possible nested entities in CTI. On the other hand, most of the existing studies focus on English CTIs, and the existing models performed poorly in a limited number of Chinese CTI studies. Given the above challenges, we propose in this paper a novel unified model, RBTG, which aims to identify flat and nested entities in Chinese CTI effectively. To overcome the difficult boundary recognition problem and the direction-dependent and distance-dependent properties in Chinese CTI NER, we use Global Pointer as the decoder and TENER as the encoder layer, respectively. Specifically, the Global Pointer layer solves the problem of the insensitivity of general NER methods to entity boundaries by utilizing the relative position information and the multiplicative attention mechanism. The TENER layer adapts to the Chinese CTI NER task by introducing an attention mechanism with direction awareness and distance awareness. Meanwhile, to cope with the complex feature capture of hierarchical structure and dependencies among Chinese CTI nested entities, the TENER layer solves the problem by following the structure of multiple self-attention layers and feed-forward network layers superimposed on each other in the Transformer. In addition, to fill the gap in the Chinese CTI nested entity dataset, we further apply the Large Language Modeling (LLM) technique and domain knowledge to construct a high-quality Chinese CTI nested entity dataset, CDTinee, which consists of six entity types selected from STIX, including nearly 4000 entity types extracted from more than 3000 threatening sentences. In the experimental session, we conduct extensive experiments on multiple datasets, and the results show that the proposed model RBTG outperforms the baseline model in both flat NER and nested NER.

Postgraduate psychological stress detection from social media using BERT-Fused model.

Postgraduate students face various academic, personal, and social stressors that increase their risk of anxiety, depression, and suicide. Identifying cost-effective methods of detecting and intervening before stress turns into severe problems is crucial. However, existing stress detection methods typically rely on psychological scales or devices, which can be complex and expensive. Therefore, we propose a BERT-fused model for rapidly and automatically detecting postgraduate students' psychological stress via social media. First, we construct an improved BERT-LDA feature extraction algorithm to extract group stress features from large-scale and complex social media data. Then, we integrate the BiLSTM-CRF named entity recognition model to construct a multi-dimensional psychological stress profile and analyze the fine-grained feature representation under the fusion of multi-dimensional features. Experimental results demonstrate that the proposed model outperforms traditional models such as BiLSTM, achieving an accuracy of 92.55%, a recall of 93.47%, and an F1-score of 92.18%, with F1-scores exceeding 89% for all three types of entities. This research provides both theoretical and practical foundations for universities or institutions to conduct fine-grained perception and intervention for postgraduate students' psychological stress.

A Communicational Ontology Inspired by Peter Singer

ABSTRACT The article states that the communicational world consists of four types of entities: 1) those that can suffer but cannot be held responsible for their communicative behavior (e.g. babies and some animals), 2) those that can suffer and can be held responsible for their communicative actions (journalists, teachers, entertainers), 3) those entities that cannot suffer but can be held responsible for their communicative behavior (media organizations, communication technologies, journalism as an institution), and 4) those entities that need to be acted on to minimize the suffering of others (climate, journalism, and potentially all entities). This “theory of moral-communicational status” explains what underlies the needs and obligations of all entities. In particular, it implies that animals should be better taken into account in communication and media ethics.

Aviation-BERT-NER: Named Entity Recognition for Aviation Safety Reports

This work introduces Aviation-BERT-NER, a Named Entity Recognition (NER) system tailored for aviation safety reports, building on the Aviation-BERT base model developed at the Georgia Institute of Technology’s Aerospace Systems Design Laboratory. This system integrates aviation domain-specific data, including aircraft types, manufacturers, quantities, and aviation terminology, to identify named entities critical for aviation safety analysis. A key innovation of Aviation-BERT-NER is its template-based approach to fine-tuning, which utilizes structured datasets to generate synthetic training data that mirror the complexity of real-world aviation safety reports. This method significantly improves the model’s generalizability and adaptability, enabling rapid updates and customization to meet evolving domain-specific requirements. The development process involved careful data preparation, including the synthesis of entity types and the generation of labeled datasets through template filling. Testing on real-world narratives from the National Transportation Safety Board (NTSB) database highlighted Aviation-BERT-NER’s robustness, with a precision of 95.34%, recall of 94.62%, and F1 score of 94.78% when evaluated over 50 manually annotated (BIO tagged) paragraphs. This work addresses a critical gap in English language NER models for aviation safety, promising substantial improvements in the analysis and understanding of aviation safety reports.

Dual Contrastive Learning for Cross-Domain Named Entity Recognition

Benefiting many information retrieval applications, named entity recognition (NER) has shown impressive progress. Recently, there has been a growing trend to decompose complex NER tasks into two subtasks (e.g., entity span detection (ESD) and entity type classification (ETC), to achieve better performance. Despite the remarkable success, from the perspective of representation, existing methods do not explicitly distinguish non-entities and entities, which may lead to ESD errors. Meanwhile, they do not explicitly distinguish entities with different entity types, which may lead to entity type misclassification. As such, the limited representation abilities may challenge some competitive NER methods, leading to unsatisfactory performance, especially in the low-resource setting (e.g., cross-domain NER). In light of these challenges, we propose to utilize contrastive learning to refine the original chaotic representations and learn the generalized representations for cross-domain NER. In particular, this article proposes a dual contrastive learning model (Dual-CL), which respectively utilizes a token-level contrastive learning module and a sentence-level contrastive learning module to enhance ESD, ETC for cross-domain NER. Empirical results on 10 domain pairs under two different settings show that Dual-CL achieves better performances than compared baselines in terms of several standard metrics. Moreover, we conduct detailed analyses to are presented to better understand each component’s effectiveness.

EvoPath: Evolutionary meta-path discovery with large language models for complex heterogeneous information networks

Heterogeneous Information Networks (HINs) encapsulate diverse entity and relation types, with meta-paths providing essential meta-level semantics for knowledge reasoning, although their utility is constrained by discovery challenges. While Large Language Models (LLMs) offer new prospects for meta-path discovery due to their extensive knowledge encoding and efficiency, their adaptation faces challenges such as corpora bias, lexical discrepancies, and hallucination. This paper pioneers the mitigation of these challenges by presenting EvoPath, an innovative framework that leverages LLMs to efficiently identify high-quality meta-paths. EvoPath is carefully designed, with each component aimed at addressing issues that could lead to potential knowledge conflicts. With a minimal subset of HIN facts, EvoPath iteratively generates and evolves meta-paths by dynamically replaying meta-paths in the buffer with prioritization based on their scores. Comprehensive experiments on three large, complex HINs with hundreds of relations demonstrate that our framework, EvoPath, enables LLMs to generate high-quality meta-paths through effective prompting, confirming its superior performance in HIN reasoning tasks. Further ablation studies validate the effectiveness of each module within the framework.

Types of legal entities under public law in German doctrine and legislation

This article examines the concepts and types of legal entities under public law in German doctrine and legislation. The legal status of public legal entities is regulated by the provisions of the German Constitution, the German Civil Code, the Law on Administrative Procedures, the Law «On the Procedure for Consideration of Cases Under the Management of Administrative Bodies», the Law «On Administrative Courts». The author examines the issue of determining the criteria for the classification of legal entities, so legal entities under public law are created in an executive order, act to satisfy the public or public interest, have authority, operate in organizational and legal forms of an open type, are responsible for the activities of their officials, legal the regime of property has certain features, in particular, such property can be transferred for use by legal entities under public law. As a rule, legal entities under public law have a special scope of legal capacity, unlike legal entities under private law, in this regard, the article analyzes the doctrine of ultra vires. The author examines the procedure for creating legal entities under public law in Germany, the types of organizational and legal forms of legal entities under public law that carry out their activities in the form of a public corporation, public institution, public fund. The article considers the sources of property formation of legal entities under public law, in particular, it can be income from entrepreneurial activity, budget funds, contributions of members of public corporations, donations (for example, the German State Bank, the German Chamber of Commerce and Industry, the Prussian Cultural Heritage Fund). The article considers four types of public corporations: territorial, united, personal, real. A feature of the activity of public corporations in Germany is the presence of membership, unlike public institutions.

M&A contracts in Ukraine: legal regulation and the mechanism of their conclusion

The article is devoted to the study of legal regulation of M&A contracts and mechanisms of their conclusion in Ukraine as a generally accepted way of business scaling. The author establishes that, according to the current regulatory provisions, M&A contracts may be embodied in the following ways: reorganisation procedures of mergers and acquisitions; establishment of control over an enterprise by acquiring a share in the authorised capital or shares of a company; asset deal, as well as a loan agreement with an alternative obligation within the Diia City legal regime. The author stated the diversity of legislation on reorganisation in the form of mergers and acquisitions - the detailed, step-by-step, European-compliant regulation of mergers and acquisitions of joint stock companies contrasts with fragmented and scattered legislation on reorganisation of other types of business entities. It has been emphasised that despite the complexity and resource­consuming nature of reorganisation in the form of mergers and acquisitions, business entities use these procedures due to their inherent universal succession, which in turn leads to tax-neutral consequences of the changes. The author additionally argues that negotiability relates to the share in the authorised capital of a company, and not to corporate rights. Based on the analysis of the legislative framework and law enforcement practice, the author establishes that an alienation agreement of a share in the authorised capital may be: concluded with both company members and third parties; with or without consideration; concluded only in respect of a paid share (at the same time, the legislative prohibition on alienation of unpaid shares in an LLC is aimed at protecting the legitimate interests of other persons, not company members who have not paid for the share); concluded both orally and in writing. The author outlines the regulatory peculiarities of acquiring a significant and controlling shareholding in joint stock companies. The author systematises the types of contracts which are asset deals by their very nature. It has been substantiated the expediency of transferring debts as a part of a single property complex to its acquirer. The article examines the legal regulation of loan agreements with alternative obligations, which are in essence analogous to the convertible loan mechanism common abroad. The author emphasises that despite the principle of freedom of contract, a loan agreement with an alternative obligation cannot be used outside the legal regime of the Diia City. It was concluded that it is necessary to regulate in more detail the conclusion of such an agreement; the variables used in calculating the creditor’s share; the rights and obligations of the creditor and the debtor.

Features of the regulation of the legal status of public law legal entities in France

In the article, the author considers issues related to the legal status of legal entities under public law in France. The legal status of legal entities under public law is regulated by the Constitution of France, special legislation in the field of administrative law. The article analyzes the criteria for distinguishing legal entities into legal entities of private and public law, developed by the judicial practice of French courts. In the article, the author examines the types of legal entities of French public law, in particular, the latter include: the state, territorial entities, public institutions. The article examines the legal nature and types of public service, which is carried out in the field of defense, transport, and industry management. The article defines a public service, as a rule, the specified services do not have the status of a legal entity, the specified status belongs to the state that creates these services. The author researches public institutions of France that belong to legal entities under public law. The article defines the concept of a public institution, classifies public institutions according to various criteria, and defines the organizational structure of these institutions. A variety of public institutions is industrial and commercial in nature. The author examines the historical aspects of the emergence and development of this type of institution. The article examines the characteristics of a public institution, in particular, the following characteristics are distinguished: 1) it is a legal entity under public law; 2) has a socially beneficial purpose; 3) is under the supervision (guardianship) of the state can be fully applied to the legal regulation of public benefit institutions, which also have the status of a legal entity, pursue a socially beneficial goal, and are under the supervision (guardianship) of the state. The article examines the legal nature of associations of persons of liberal professions (lawyers, attorneys, notaries, doctors, pharmacists, veterinarians, and others). The article examines the judicial practice of French courts on issues of compensation for damage caused by the activities of a state institution. The author examines and investigates issues of the legal regime of property of legal entities under public law.

Use of SNOMED CT in Large Language Models: Scoping Review.

Large language models (LLMs) have substantially advanced natural language processing (NLP) capabilities but often struggle with knowledge-driven tasks in specialized domains such as biomedicine. Integrating biomedical knowledge sources such as SNOMED CT into LLMs may enhance their performance on biomedical tasks. However, the methodologies and effectiveness of incorporating SNOMED CT into LLMs have not been systematically reviewed. This scoping review aims to examine how SNOMED CT is integrated into LLMs, focusing on (1) the types and components of LLMs being integrated with SNOMED CT, (2) which contents of SNOMED CT are being integrated, and (3) whether this integration improves LLM performance on NLP tasks. Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines, we searched ACM Digital Library, ACL Anthology, IEEE Xplore, PubMed, and Embase for relevant studies published from 2018 to 2023. Studies were included if they incorporated SNOMED CT into LLM pipelines for natural language understanding or generation tasks. Data on LLM types, SNOMED CT integration methods, end tasks, and performance metrics were extracted and synthesized. The review included 37 studies. Bidirectional Encoder Representations from Transformers and its biomedical variants were the most commonly used LLMs. Three main approaches for integrating SNOMED CT were identified: (1) incorporating SNOMED CT into LLM inputs (28/37, 76%), primarily using concept descriptions to expand training corpora; (2) integrating SNOMED CT into additional fusion modules (5/37, 14%); and (3) using SNOMED CT as an external knowledge retriever during inference (5/37, 14%). The most frequent end task was medical concept normalization (15/37, 41%), followed by entity extraction or typing and classification. While most studies (17/19, 89%) reported performance improvements after SNOMED CT integration, only a small fraction (19/37, 51%) provided direct comparisons. The reported gains varied widely across different metrics and tasks, ranging from 0.87% to 131.66%. However, some studies showed either no improvement or a decline in certain performance metrics. This review demonstrates diverse approaches for integrating SNOMED CT into LLMs, with a focus on using concept descriptions to enhance biomedical language understanding and generation. While the results suggest potential benefits of SNOMED CT integration, the lack of standardized evaluation methods and comprehensive performance reporting hinders definitive conclusions about its effectiveness. Future research should prioritize consistent reporting of performance comparisons and explore more sophisticated methods for incorporating SNOMED CT's relational structure into LLMs. In addition, the biomedical NLP community should develop standardized evaluation frameworks to better assess the impact of ontology integration on LLM performance.

Named Entity Recognition for Equipment Fault Diagnosis Based on RoBERTa-wwm-ext and Deep Learning Integration

Equipment fault diagnosis NER is to extract specific entities from Chinese equipment fault diagnosis text, which is the premise of constructing an equipment fault diagnosis knowledge graph. Named entity recognition for equipment fault diagnosis can also provide important data support for equipment maintenance support. Equipment fault diagnosis text has complex semantics, fuzzy entity boundaries, and limited data size. In order to extract entities from the equipment fault diagnosis text, this paper presents an NER model for equipment fault diagnosis based on RoBERTa-wwm-ext and Deep Learning network integration. Firstly, this model uses the RoBERTa-wwm-ext to extract context-sensitive embeddings of text sequences. Secondly, the context feature information is obtained through the BiLSTM network. Thirdly, the CRF is combined to output the label sequence with a constraint relationship, improve the accuracy of sequence labeling task, and complete the entity recognition task. Finally, experiments and predictions are carried out on the constructed dataset. The results show that the model can effectively identify five types of equipment fault diagnosis entities and has higher evaluation indexes than the traditional model. Its precision, recall, and F1 value are 94.57%, 95.39%, and 94.98%, respectively. The case study proves that the model can accurately recognize the entity of the input text.

Augmenting biomedical named entity recognition with general-domain resources

ObjectiveTraining a neural network-based biomedical named entity recognition (BioNER) model usually requires extensive and costly human annotations. While several studies have employed multi-task learning with multiple BioNER datasets to reduce human effort, this approach does not consistently yield performance improvements and may introduce label ambiguity in different biomedical corpora. We aim to tackle those challenges through transfer learning from easily accessible resources with fewer concept overlaps with biomedical datasets. MethodsWe proposed GERBERA, a simple-yet-effective method that utilized general-domain NER datasets for training. We performed multi-task learning to train a pre-trained biomedical language model with both the target BioNER dataset and the general-domain dataset. Subsequently, we fine-tuned the models specifically for the BioNER dataset. ResultsWe systematically evaluated GERBERA on five datasets of eight entity types, collectively consisting of 81,410 instances. Despite using fewer biomedical resources, our models demonstrated superior performance compared to baseline models trained with additional BioNER datasets. Specifically, our models consistently outperformed the baseline models in six out of eight entity types, achieving an average improvement of 0.9% over the best baseline performance across eight entities. Our method was especially effective in amplifying performance on BioNER datasets characterized by limited data, with a 4.7% improvement in F1 scores on the JNLPBA-RNA dataset. ConclusionThis study introduces a new training method that leverages cost-effective general-domain NER datasets to augment BioNER models. This approach significantly improves BioNER model performance, making it a valuable asset for scenarios with scarce or costly biomedical datasets. We make data, codes, and models publicly available via https://github.com/qingyu-qc/bioner_gerbera.

Uncovering condition information loss in medical text extraction: The challenge of non-contiguous spans

We investigated the limitations of conventional named entity recognition (NER) and entity linking (EL) methods in accurately extracting patient condition information from medical texts, focusing on the challenges posed by non-contiguous spans and the potential information loss. We utilized a corpus with entity-relation annotations, analyzing the frequency and nature of non-contiguous spans that include irrelevant entities within gaps. The corpus was further analyzed to pinpoint the types of entity representations predominantly linked with peripheral spans—those not encompassing central symptom-describing terms—with a focus on items, body parts, and clinical tests. Our analysis revealed that 18.6 % of patient condition expressions were non-contiguous spans containing irrelevant entities, suggesting an accuracy ceiling of 81.4 % for conventional NER and EL approaches in the worst-case scenario. The study highlights the importance of entity types such as items, body parts, and clinical tests in these expressions, indicating that conventional extraction methods incur considerable information loss. The findings underscore the need for more sophisticated information extraction techniques capable of handling the complexities of medical texts, including non-contiguous spans. Adapting methods that allow gaps within entities or employing graph-based term assignments can enhance the accuracy and comprehensiveness of medical text annotation.