A5078133624- Biomedical Text Mining and Ontologies
- Semantic Web and Ontologies
- Topic Modeling
- Scientific Computing and Data Management
- Advanced Graph Neural Networks
- Data Quality and Management
- Natural Language Processing Techniques
- Research Data Management Practices
- Advanced Text Analysis Techniques
- E-Government and Public Services
- Advanced Data Processing Techniques
- Web Data Mining and Analysis
- Artificial Intelligence in Healthcare
- Radiomics and Machine Learning in Medical Imaging
- Power Systems and Technologies
L3S Research Center
2019-2025
Leibniz University Hannover
2019-2023
Technische Informationsbibliothek (TIB)
2020
University of Bonn
2017
Despite improved digital access to scholarly knowledge in recent decades, communication remains exclusively document-based. In this form, is hard process automatically. We present the first steps towards a graph based infrastructure that acquires machine actionable form thus enabling new possibilities for curation, publication and processing. The primary contribution present, evaluate discuss multi-modal acquisition, combining crowdsourced automated techniques. results of user evaluation...
The Open Research Knowledge Graph is an infrastructure for the production, curation, publication and use of FAIR scientific information. Its mission to shape a future scholarly publishing communication where contents articles are research data.
Abstract The transfer of knowledge has not changed fundamentally for many hundreds years: It is usually document-based-formerly printed on paper as a classic essay and nowadays PDF. With around 2.5 million new research contributions every year, researchers drown in flood pseudo-digitized PDF publications. As result seriously weakened. In this article, we argue representing scholarly structured semantic way graph. advantage that information represented graph readable by machines humans. an...
Abstract In the last decade, a large number of knowledge graph (KG) completion approaches were proposed. Albeit effective, these efforts are disjoint, and their collective strengths weaknesses in effective KG have not been studied literature. We extend Plumber , framework that brings together research community’s disjoint on completion. include more components into architecture to comprise 40 reusable for various subtasks, such as coreference resolution, entity linking, relation extraction....
Knowledge graphs have gained increasing popularity in the last decade science and technology. However, knowledge are currently relatively simple to moderate semantic structures that mainly a collection of factual statements. Question answering (QA) benchmarks systems were so far geared towards encyclopedic such as DBpedia Wikidata. We present SciQA scientific QA benchmark for scholarly knowledge. The leverages Open Research Graph (ORKG) which includes almost 170,000 resources describing...
Transforming natural language questions into formal queries is an integral task in Question Answering (QA) systems. QA systems built on knowledge graphs like DBpedia, require a step after processing for linking words, specifically including named entities and relations, to their corresponding graph. To achieve this task, several approaches rely background bases containing semantically-typed e.g., PATTY, extra disambiguation step. Two major factors may affect the performance of relation...
Reviewing scientific literature is a cumbersome, time consuming but crucial activity in research. Leveraging scholarly knowledge graph, we present methodology and system for comparing literature, particular research contributions describing the addressed problem, utilized materials, employed methods yielded results. The can be used by researchers to quickly get familiar with existing work specific domain (e.g., concrete question or hypothesis). Additionally, it publish surveys following FAIR...
Literature is the primary expression of scientific knowledge and an important source research data. However, expressed in narrative text documents not inherently machine reusable. To facilitate reuse, e.g. for synthesis research, must be extracted from articles organized into databases post-publication. The high time costs inaccuracies associated with completing these activities manually has driven development techniques that automate extraction. Tackling problem a different mindset, we...
Scholarly Knowledge Graphs (KGs) provide a rich source of structured information representing knowledge encoded in scientific publications. With the sheer volume published literature comprising plethora inhomogeneous entities and relations to describe concepts, these KGs are inherently incomplete. We present exBERT, method for leveraging pre-trained transformer language models perform scholarly graph completion. model triples as text triple classification (i.e., belongs KG or not). The...
We introduce the STEM (Science, Technology, Engineering, and Medicine) Dataset for Scientific Entity Extraction, Classification, Resolution, version 1.0 (STEM-ECR v1.0). The STEM-ECR v1.0 dataset has been developed to provide a benchmark evaluation of scientific entity extraction, classification, resolution tasks in domain-independent fashion. It comprises abstracts 10 disciplines that were found be most prolific ones on major publishing platform. describe creation such multidisciplinary...
Purpose: Finding scholarly articles is a time-consuming and cumbersome activity, yet crucial for conducting science. Due to the growing number of articles, new search systems are needed effectively assist researchers in finding relevant literature. Methodology: We take neuro-symbolic approach exploration by leveraging state-of-the-art components, including semantic search, Large Language Models (LLMs), Knowledge Graphs (KGs). The component composes set articles. From this information...
Information Extraction (IE) tasks are commonly studied topics in various domains of research. Hence, the community continuously produces multiple techniques, solutions, and tools to perform such tasks. However, running those integrating them within existing infrastructure requires time, expertise, resources. One pertinent task here is triples extraction linking, where structured extracted from a text aligned an Knowledge Graph (KG). In this paper, we present Plumber , first framework that...