Hierarchical multi-label classification is a particularly difficult type of classification task. Models must handle high levels of class imbalance while adhering to the label hierarchy. One approach to this problem type are heterogeneous ensembles that combine different base learner architectures to deal with the task's varying complexities. We propose a novel architecture for heterogeneous ensembles that employs a multi-layered weighting mechanism as well as a specialised inconsistency resolution algorithm. In a case study on the Chemical Entities of Biological Interest (ChEBI) ontology, we construct an ensemble of deep learning and symbolic models, demonstrating how the proposed ensemble can be used to integrate heterogeneous base learners. In total, the ensemble covers 1,742 ChEBI classes. The proposed weighting mechanism can significantly improve classification performance compared to other aggregation strategies.

We study modular acceptability semantics for bipolar weighted argumentation graphs, which cleanly separate the aggregation of attacking and supporting arguments from the computation of their influence on an argument’s initial weight. Using a matrix representation, we reformulate the desirable characteristics of such semantics and prove general convergence and divergence theorems. We also introduce the first semantics that converges on all bipolar weighted graphs.

Modern biochemistry is producing vast amounts of chemical knowledge. Ontologies, such as the Chemical Entities of Biological Interest (ChEBI) ontology, can help organising this knowledge. With manual classification alone however, ontologies cannot keep up with the growth of their domain. In this work, we propose a novel taxonomy of 67 classes related to peptides, a large branch in ChEBI with nearly 15,000 compounds. The existing natural language definitions in ChEBI have been expanded and specified more precisely. These natural language definitions are accompanied by a logical axiomatisation in monadic second-order logic (MSOL). To use the axiomatisation for automated classification, a methodology has been developed that translates monadic second-order definitions first into partial first-order definitions and finally into an algorithmic classification. This connects three aspects important to ontological definitions: They reflect the opinions of experts, they are unambiguous, and they can be checked automatically. In our evaluation, we compare the results of our classification to the current taxonomy of ChEBI . This reveals potential inconsistencies in ChEBI as well as areas that might benefit from automated extensions. We also evaluate our natural-language definitions in an expert survey. Scientific contribution: This work provides precise natural-language definitions of 14 current ChEBI classes as well as 53 new peptide-related classes. These definitions are formalised in MSOL and come with an efficient implementation that allows for large-scale molecule classification, including a full classification of ChEBI and PubChem.

Materials Science and Engineering (MSE) increasingly relies on data‐intensive, automated, and distributed workflows that span synthesis, manufacturing, characterization, design, and simulation. These settings require machine‐actionable representations of materials and processes that remain interoperable across laboratories, software stacks, and organizations. Therefore, Platform MaterialDigital Core Ontology (PMDco) 3.0 is introduced as a mid‐level ontology that provides a semantic framework for the processing–structure–properties paradigm in MSE. PMDco 3.0 adopts an architecture aligned with the Basic Formal Ontology that enables a logically consistent classification of fundamental MSE concepts and the explicit representation of intrinsic material properties, contextual roles and functions, and related information artifacts. The work outlines the technical curation approach that supports sustainable ontology evolution through reproducible builds, automated release generation, and systematic validation workflows. Representative semantic patterns are presented as reusable building blocks for consistent modeling and data mapping, including material object duality, intensive versus extensive qualities, role and function assignment, immaterial entities for spatial context, process modeling across production, assay, and computation, and the separation of requirements from observations via set points and measurements. PMDco 3.0 is intended to serve as a community‐driven anchor for interoperable domain and application ontologies and scalable semantic interoperability in MSE.

In our work, we systematize and analyze implicit ontological commitments in the responses generated by large language models (LLMs), focusing on ChatGPT 3.5 as a case study. We investigate how LLMs exhibit implicit ontological categorizations reflected in the texts they generate, despite having no explicit ontology. The article proposes an approach to understanding the ontological commitments of LLMs by defining ontology as a theory that provides a systematic account of the ontological commitments of some text. We investigate the ontological assumptions of ChatGPT and present a systematized account, that is, GPT’s top-level ontology. This includes a taxonomy, which is available as an OWL file, as well as a discussion about ontological assumptions (e.g., about its mereology or presentism). We show that in some aspects GPT’s top-level ontology is quite similar to existing top-level ontologies. However, significant challenges arise from the flexible nature of LLM-generated texts, including ontological overload, ambiguity, and inconsistency.

In ontology development, there is a gap between domain ontologies which mostly use the Web Ontology Language, OWL, and foundational ontologies written in first-order logic, FOL. To bridge this gap, we present Gavel, a tool that supports the development of heterogeneous ‘FOWL’ ontologies that extend OWL with FOL annotations, and is able to reason over the combined set of axioms. Since FOL annotations are stored in OWL annotations, FOWL ontologies remain compatible with the existing OWL infrastructure. We show that for the OWL domain ontology OBI, the stronger integration with its FOL top-level ontology BFO via our approach enables us to detect several inconsistencies. Furthermore, existing OWL ontologies can benefit from FOL annotations. We illustrate this with FOWL ontologies containing mereotopological axioms that enable additional, useful inferences. Finally, we show that even for large domain ontologies such as ChEBI, automatic reasoning with FOL annotations can be used to detect previously unnoticed errors in the classification.

Accurately classifying chemical structures is a central task in cheminformatics. Different approaches to classification exist, but they all come with their own drawbacks. Manual classification is time-intensive and hard to scale up to large databases. Rule-based methods are specific to some classes and require a high level of maintenance. Deep learning models lack reliability and explainability. In this work, we present an ensemble learning method that combines generative artificial intelligence, classical deep learning and symbolic approaches to classify chemicals into the Chemical Entities of Biological Interest (ChEBI) ontology. Our approach covers 1,722 classes with vastly different properties. The ensemble makes use of the strengths of each model and aligns their predictions with the OWL axiomatisation of ChEBI. We provide both a Python library as well as a web front-end which allow users to classify arbitrary molecules with the ensemble.

Despite its prevalence, in many domains, OWL is not expressive enough to define ontology classes. In this paper, we present an approach that allows to use monadic second-order formalisations for ontology classification. As a case study, we have applied our approach to 14 peptide-related classes from the chemistry ontology ChEBI. For these classes, a monadic second-order logic formalisation has been developed and applied both to ChEBI as well as to 119 million molecules from the chemistry database PubChem. While this logical approach alone is limited to classification for the specified classes (in our case, (sub)classes of peptides), transformer deep learning models scale classification to the whole of the ChEBI ontology. We show that when using the classifications obtained by the logical approach as training data, the performance of the deep learning models can be significantly enhanced.

Recent advances in the field of machine learning have shown that integration of expert knowledge improves performances, in particular for complex domains such as biology. Bio-ontologies offer a rich source of curated biological knowledge that can be harnessed to this end. Here, we describe an intuitive and generalisable approach to embed the knowledge contained in a classification hierarchy derived from a bio-ontology into a machine learning model as an intermediate training step between general-purpose pre-training and task-specific fine-tuning in a process that we call ‘ontology pre-training’. We show that this approach leads to an improvement in predictive performance and a reduction in training time for a broad range of prediction tasks relevant to understanding metabolite functions in living systems, using a range of datasets derived from MoleculeNet. We see the biggest improvement for regression tasks, e.g. prediction of lipophilicity and aqueous solubility of molecules, and a robust improvement for most classification tasks. Our approach can be adapted for a wide range of knowledge sources, models and prediction tasks. ### Competing Interest Statement The authors have declared no competing interest. Swiss National Science Foundation, 215906 Deutsche Forschungsgemeinschaft, https://ror.org/018mejw64, 522907718

Ontologies often involve complex logical structures, so changes to individual classes or the addition of new axioms can have significant implications for other parts of the ontology. Due to this complexity, dependencies between symbols in the vocabulary of the ontology are not always immediately apparent. In this paper, we define three semantics-based approaches for establishing dependency relationships between these symbols and explore their specific properties. Additionally, we apply these dependency relations in a case study.

Complex systems are full of unpredictable and uncertain behaviours that depend on many factors. Various fields of science have set themselves the task of studying these systems and predicting their behaviour under various premises with the help of computer models. Kwakkel et al. [1] published an uncertainty matrix to systematically record and communicate uncertainties about systems. This paper takes up the matrix, analyses the concepts of model and uncertainty in literature, and models them ontologically in the Modelling Uncertainties Ontology (MUNO), a BFO-based mid-level ontology for modelling uncertainties in models for many domains. MUNO is evaluated against requirements and competency questions. Based on the ontology, an RDF-shape is created and used as an example in a case study to annotate the uncertainties for an energy system model.

BFO 2020 does not support functions, dispositions, and roles of generically dependent continuants (like software or datasets). In this paper, we argue that this is a severe limitation, which prevents, for example, the adequate representation of the functions of computer models or the various roles of datasets during the execution of these models. We discuss the aspects of BFO 2020 that prevent the representation of realizable entities of generically dependent continuants. Two approaches to address the issue are presented: (a) the use of defined classes and (b) a proposal of changes that allow BFO to support functions, dispositions, and roles of generically dependent continuants.

Deriving symbolic knowledge from trained deep learning models is challenging due to the lack of transparency in such models. A promising approach to address this issue is to couple a semantic structure with the model outputs and thereby make the model interpretable. In prediction tasks such as multi-label classification, labels tend to form hierarchical relationships. Therefore, we propose enforcing a taxonomical structure on the model’s outputs throughout the training phase. In vector space, a taxonomy can be represented using axis-aligned hyper-rectangles, or boxes, which may overlap or nest within one another. The boundaries of a box determine the extent of a particular category. Thus, we used box-shaped embeddings of ontology classes to learn and transparently represent logical relationships that are only implicit in multi-label datasets. We assessed our model by measuring its ability to approximate the full set of inferred subclass relations in the ChEBI ontology, which is an important knowledge base in the field of life science. We demonstrate that our model captures implicit hierarchical relationships among labels, ensuring consistency with the underlying ontological conceptualization, while also achieving state-of-the-art performance in multi-label classification. Notably, this is accomplished without requiring an explicit taxonomy during the training process. Our proposed approach advances chemical classification by enabling interpretable outputs through a structured and geometrically expressive representation of molecules and their classes.

Reference ontologies provide a shared vocabulary and knowledge resource for their domain. Manual construction and annotation enables them to maintain high quality, allowing them to be widely accepted across their community. However, the manual ontology development process does not scale for large domains. We present a new methodology for automatic ontology extension for domains in which the ontology classes have associated graph-structured annotations, and apply it to the ChEBI ontology, a prominent reference ontology for life sciences chemistry. We train Transformer-based deep learning models on the leaf node structures from the ChEBI ontology and the classes to which they belong. The models are then able to automatically classify previously unseen chemical structures, resulting in automated ontology extension. The proposed models achieved an overall F1 scores of 0.80 and above, improvements of at least 6 percentage points over our previous results on the same dataset. In addition, the models are interpretable: we illustrate that visualizing the model’s attention weights can help to explain the results by providing insight into how the model made its decisions. We also analyse the performance for molecules that have not been part of the ontology and evaluate the logical correctness of the resulting extension.

Deep learning models are often unaware of the inherent constraints of the task they are applied to. However, many downstream tasks require logical consistency. For ontology classification tasks, such constraints include subsumption and disjointness relations between classes. In order to increase the consistency of deep learning models, we propose a fuzzy loss that combines label-based loss with terms penalising subsumption- or disjointness-violations. Our evaluation on the ChEBI ontology shows that the fuzzy loss is able to decrease the number of consistency violations by several orders of magnitude without decreasing the classification performance. In addition, we use the fuzzy loss for unsupervised learning. We show that this can further improve consistency on data from a distribution outside the scope of the supervised training.

Ontologies are formal representations of knowledge in specific domains that provide a structured framework for organizing and understanding complex information. Creating ontologies, however, is a complex and time-consuming endeavor. ChEBI is a well-known ontology in the field of chemistry, which provides a comprehensive resource for defining chemical entities and their properties. However, it covers only a small fraction of the rapidly growing knowledge in chemistry and does not provide references to the scientific literature. To address this, we propose a methodology that involves augmenting existing annotated text corpora with knowledge from Chebi and fine-tuning a large language model (LLM) to recognize chemical entities and their roles in scientific text. Our experiments demonstrate the effectiveness of our approach. By combining ontological knowledge and the language understanding capabilities of LLMs, we achieve high precision and recall rates in identifying both the chemical entities and roles in scientific literature. Furthermore, we extract them from a set of 8,000 ChemRxiv articles, and apply a second LLM to create a knowledge graph (KG) of chemical entities and roles (CEAR), which provides complementary information to ChEBI, and can help to extend it.

Energy is a fundamental phenomenon of physics, but energy also plays an important role in the representation of many domains, since many processes involve energy transformation or transfer. However, energy is represented very differently in existing ontologies. Even in ontologies that share BFO as top-level ontology, energy is sometimes treated as disposition, as quality, and as a material entity. As we discuss in the paper, there are reasons for each choice, which makes the ontological representation of energy a challenging subject. In this paper we present an ontological analysis of energy in a BFO-based mid-level ontology (MENO), including the different kinds of energy, their relations to dispositions as well as their realisation in processes.

Connecting chemical structural representations with meaningful categories and semantic annotations representing existing knowledge enables data-driven digital discovery from chemistry data. Ontologies are semantic annotation resources that provide definitions and a classification hierarchy for a domain. They are widely used throughout the life sciences. ChEBI is a large-scale ontology for the domain of biologically interesting chemistry that connects representations of chemical structures with meaningful chemical and biological categories. Classifying novel molecular structures into ontologies such as ChEBI has been a longstanding objective for data scientific methods, but the approaches that have been developed to date are limited in several ways: they are not able to expand as the ontology expands without manual intervention, and they are not able to learn from continuously expanding data. We have developed an approach for automated classification of chemicals in the ChEBI ontology based on a neuro-symbolic AI technique that harnesses the ontology itself to create the learning system. We provide this system as a publicly available tool, Chebifier, and as an API, ChEB-AI. We here evaluate our approach and show how it constitutes an advance towards a continuously learning semantic system for chemical knowledge discovery.

Image schemas are mental patterns learned from perceptual experiences capturing conceptual constructions in expressions. In linguistic analysis, their visualizations are often context-dependent without a generalizable structure. Addressing this, we introduce The Diagrammatic Image Schema Language: a formal representation language that systematizes a set of visual combination rules for different conceptual primitives from the cognitive science literature. These primitives are distinguished from a formal point of view to allow for more general application. DISL also contains a logical exchange format in which the diagrams may be made machine-readable. Using DISL, the semantic structure of complex scenarios can be represented and computed.