Arrow Research search

Author name cluster

Frank van Harmelen

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

34 papers
2 author rows

Possible papers

34

JBHI Journal 2025 Journal Article

KG4NH: A Comprehensive Knowledge Graph for Question Answering in Dietary Nutrition and Human Health

  • Chengcheng Fu
  • Xueli Pan
  • Jieyu Wu
  • Junkai Cai
  • Zhisheng Huang
  • Frank Van Harmelen
  • Weizhong Zhao
  • Xingpeng Jiang

It is commonly known that food nutrition is closely related to human health. The complex interactions between food nutrients and diseases, influenced by gut microbial metabolism, present challenges in systematizing and practically applying knowledge. To address this, we propose a method for extracting triples from a vast amount of literature, which is used to construct a comprehensive knowledge graph on nutrition and human health. Concurrently, we develop a query-based question answering system over our knowledge graph, proficiently addressing three types of questions. The results show that our proposed model outperforms other state-of-art methods, achieving a precision of 0. 92, a recall of 0. 81, and an F1 score of 0. 86 in the nutrition and disease relation extraction task. Meanwhile, our question answering system achieves an accuracy of 0. 68 and an F1 score of 0. 61 on our benchmark dataset, showcasing competitiveness in practical scenarios. Furthermore, we design five independent experiments to assess the quality of the data structure in the knowledge graph, ensuring results characterized by high accuracy and interpretability. In conclusion, the construction of our knowledge graph shows significant promise in facilitating diet recommendations, enhancing patient care applications, and informing decision-making in clinical research.

Details DOI

NAI Journal 2025 Journal Article

Towards semantically enriched embeddings for knowledge graph completion

  • Mehwish Alam
  • Frank Van Harmelen
  • Maribel Acosta

Embedding based Knowledge Graph (KG) completion has gained much attention over the past few years. Most of the current algorithms consider a KG as a multidirectional labeled graph and lack the ability to capture the semantics underlying the schematic information. This position paper revises the state of the art and discusses several variations of the existing algorithms for KG completion, which are discussed progressively based on the level of expressivity of the semantics utilized. The paper begins with analysing various KG completion algorithms considering only factual information such as transductive and inductive link prediction and entity type prediction algorithms. It then revises the algorithms utilizing Large Language Models as background knowledge. Afterwards, it discusses the algorithms progressively utilizing semantic information such as class hierarchy information within the KGs and semantics represented in different description logic axioms. The paper concludes with a critical reflection on the current state of work in the community, where we argue that the aspects of semantics, rigorous evaluation protocols, and bias against external sources have not been sufficiently addressed in the literature, which hampers a more thorough understanding of advantages and limitations of existing approaches. Lastly, we provide recommendations for future directions.

Details DOI

NeurIPS Conference 2023 Conference Paper

A-NeSI: A Scalable Approximate Method for Probabilistic Neurosymbolic Inference

  • Emile van Krieken
  • Thiviyan Thanapalasingam
  • Jakub Tomczak
  • Frank Van Harmelen
  • Annette Ten Teije

We study the problem of combining neural networks with symbolic reasoning. Recently introduced frameworks for Probabilistic Neurosymbolic Learning (PNL), such as DeepProbLog, perform exponential-time exact inference, limiting the scalability of PNL solutions. We introduce Approximate Neurosymbolic Inference (A-NeSI): a new framework for PNL that uses neural networks for scalable approximate inference. A-NeSI 1) performs approximate inference in polynomial time without changing the semantics of probabilistic logics; 2) is trained using data generated by the background knowledge; 3) can generate symbolic explanations of predictions; and 4) can guarantee the satisfaction of logical constraints at test time, which is vital in safety-critical applications. Our experiments show that A-NeSI is the first end-to-end method to solve three neurosymbolic tasks with exponential combinatorial scaling. Finally, our experiments show that A-NeSI achieves explainability and safety without a penalty in performance.

PDF Details

IJCAI Conference 2022 Conference Paper

Reinforcement Learning with Option Machines

  • Floris den Hengst
  • Vincent Francois-Lavet
  • Mark Hoogendoorn
  • Frank Van Harmelen

Reinforcement learning (RL) is a powerful framework for learning complex behaviors, but lacks adoption in many settings due to sample size requirements. We introduce a framework for increasing sample efficiency of RL algorithms. Our approach focuses on optimizing environment rewards with high-level instructions. These are modeled as a high-level controller over temporally extended actions known as options. These options can be looped, interleaved and partially ordered with a rich language for high-level instructions. Crucially, the instructions may be underspecified in the sense that following them does not guarantee high reward in the environment. We present an algorithm for control with these so-called option machines (OMs), discuss option selection for the partially ordered case and describe an algorithm for learning with OMs. We compare our approach in zero-shot, single- and multi-task settings in an environment with fully specified and underspecified instructions. We find that OMs perform significantly better than or comparable to the state-of-art in all environments and learning settings.

PDF Details DOI

KR Conference 2020 Conference Paper

Analyzing Differentiable Fuzzy Implications

  • Emile van Krieken
  • Erman Acar
  • Frank Van Harmelen

Combining symbolic and neural approaches has gained considerable attention in the AI community, as it is argued that their strengths and weaknesses are complementary. One trend in the literature are weakly supervised learning techniques that employ operators from fuzzy logics. They use prior background knowledge described in such logics to help training neural networks from unlabeled and noisy data. By interpreting logical symbols using neural networks (or grounding them), this background knowledge can be added to regular loss functions, hence making reasoning a part of learning. We investigate how implications from the fuzzy logic literature behave in a differentiable setting. In this setting, we analyze the differences between the formal properties of these fuzzy implications. It turns out that various fuzzy implications, including some of the most well-known, are highly unsuitable for use in a differentiable learning setting. A further finding shows a strong imbalance between gradients driven by the antecedent and the consequent of the implication. Furthermore, we introduce a new family of fuzzy implications (called sigmoidal implications) to tackle this phenomenon. Finally, we empirically show that it is possible to use Differentiable Fuzzy Logics for semi-supervised learning, and show that sigmoidal implications outperform other choices of fuzzy implications.

PDF Details DOI

FLAP Journal 2019 Journal Article

Semi-supervised Learning using Differentiable Reasoning.

  • Emile van Krieken
  • Erman Acar
  • Frank Van Harmelen

We introduce Differentiable Reasoning (DR), a novel semi-supervised learning technique which uses relational background knowledge to benefit from unlabeled data. We apply it to the Semantic Image Interpretation (SII) task and show that background knowledge provides significant improvement. We find that there is a strong but interesting imbalance between the contributions of updates from Modus Ponens (MP) and its logical equivalent Modus Tollens (MT) to the learning process, suggesting that our approach is very sensitive to a phenomenon called the Raven Paradox [10]. We propose a solution to overcome this situation.

Details

IS Journal 2009 Journal Article

It's a Streaming World! Reasoning upon Rapidly Changing Information

  • Emanuele Della Valle
  • Stefano Ceri
  • Frank Van Harmelen
  • Dieter Fensel

Data streams occur in modern applications such as sensor network monitoring, traffic engineering, RFID tags applications, telecom call recording, medical record management, financial applications, and clickstreams. On the Web, many sites distribute and present information in real-time streams. In many of these application areas, the ability to perform complex reasoning tasks that combine streaming data with evolving knowledge would be of great benefit. Stream reasoning—an unexplored, yet high-impact research area—is a new multidisciplinary approach that will build on the Semantic Web and provide the abstractions, foundations, methods, and tools required to integrate data streams and reasoning systems.

Details DOI

IJCAI Conference 2005 Conference Paper

Reasoning with Inconsistent Ontologies

  • Zhisheng Huang
  • Frank Van Harmelen
  • Annette ten

In this paper we present a framework of reasoning with inconsistent ontologies, in which pre-defined selection functions are used to deal with concept relevance. We examine how the notion of ”concept relevance” can be used for reasoning with inconsistent ontologies. We have implemented a prototype called PION (Processing Inconsistent ONtologies), which is based on a syntactic relevance-based selection function. In this paper, we also report the experiments with PION.

PDF

KER Journal 2002 Journal Article

How the semantic web will change KR: challenges and opportunities for a new research agenda

  • Frank Van Harmelen

Currently the Web is the largest available environment for the deployment of agents, and much work in agent research is driven by Web-based applications (Luke et al. (1997), Joachims et al. (1997), Bollacker et al. (1998), Doorenbos et al. (1997) are just some examples; see also the May 2000 special issue of the Artificial Intelligence Journal on intelligent internet systems, 118 (1–2)). However, such applications of agent technology are hampered by the fact that the Web is not geared towards agent use, but is rather designed for human use. Current Web resources are lacking in explicit, machine-accessible descriptions of their contents; they are only fully accessible to agents with a competent grasp of English (i.e. limited to human agents only).

Details DOI

KER Journal 1995 Journal Article

Formal methods in knowledge engineering

  • Frank Van Harmelen
  • Dieter Fensel

Abstract This paper presents a general discussion of the role of formal methods in knowledge engineering. We give an historical account of the development of the field of knowledge engineering towards the use of formal methods. Subsequently, we discuss the pros and cons of formal methods. We do this by summarising the proclaimed advantages, and by arguing against some of the commonly heard objections against formal methods. We briefly summarise the current state of the art and discuss the most important directions that future research in this field should take. This paper presents a general setting for the other contributions in this issue of the journal, which each deal with a specific issue in more detail.

Details DOI

KER Journal 1994 Journal Article

A comparison of languages which operationalize and formalize KADS models of expertise

  • Dieter Fensel
  • Frank Van Harmelen

Abstract In the field of knowledge engineering, dissatisfaction with the rapid-prototyping approach has led to a number of more principled methodologies for the contruction of knowledge-based systems. Instead of immediately implementing the gathered and interpreted knowledge in a given implementation formalism according to the rapid-prototyping approach, many such methodologies centre around the notion of a conceptual model: an abstract, implementation independent description of the relevant problem solving expertise. A conceptual model should describe the task which is solved by the system and the knowledge which is required by it. Although such conceptual models more precisely, and operationally as a means for model evaluation. In this paper, we study a number of such formal and operational languages for specifying conceptual models. To enable a meaningful comparison of such languages, we focus on languages which are all aimed at the same underlying conceptual model, namely that from the KADS method for building KBS. We describe eight formal languages for KADS models of expertise, and compare these languages with respect to their modelling primitives, their semantics, their implementations and their applications, Future research issues in the area of formal and operational specification languages for KBS are identified as the result of studying these languages. The paper also contains an extensive bibliography of research in this area.

Details DOI

IJCAI Conference 1989 Conference Paper

A Rational Reconstruction and Extension of Recursion Analysis

  • Alan Bundy
  • Frank Van Harmelen
  • JANE HESKETH
  • Alan Smaill
  • Andrew Stevens

The focus of this paper is the technique of recur8\on analysis. Recursion analysis is used by the Boyer-Moore Theorem Prover to choose an appropriate induction schema and variable to prove theorems by mathematical induction. A rational reconstruction of recursion analysis is outlined, using the technique of proof plans. This rational reconstruction suggests an extension of recursion analysis which frees the induction suggestion from the forms of recursion found in the conjecture. Preliminary results are reported of the automation of this rational reconstruction and extension using the CLAM- Oyster system.

PDF

KER Journal 1984 Journal Article

Criteria for Choosing Representation Languages and Control Regimes for Expert Systems

  • Han Reichgelt
  • Frank Van Harmelen

Abstract Shells and high-level programming language environments suffer from a number of shortcomings as knowledge engineering tools. We conclude that a variety of knowledge representation formalisms and a variety of controls regimes are needed. In addition guidelines should be provided about when to choose which knowledge representation formalism and which control regime. The guidelines should be based on properties of the task and the domain of the expert system. In order to arrive at these guidelines we first critically review some of the classifications of expert systems in the literature. We then give our own list of criteria. We test this list applying our criteria to a number of existing expert systems. As a caveat, we have not yet made a systematic attempt at correlating the criteria and different knowledge representations formalisms and control regimes, although we make some preliminary remarks throughout the paper.

Details DOI