DeepOnto: A Python package for ontology engineering with deep learning

To bridge the semantic gap between different offered ontologies, Ontology Engineers have proposed and implemented various matching techniques during the last decade. Still, these techniques require more efforts and investigations to achieve better precision and recall for the set of alignments. The ontology alignment process consists of mainly three different stages; namely, String based, Linguistic based and Structure based. This paper presents the evaluation and analysis of various linguistic based ontology alignment techniques and their performance during the overall alignment process. For this purpose, we have included those linguistic matching tools which are widely used in the alignment systems whose performance is relatively better in terms of precision and recall during the tests conducted for Ontology Alignment Evaluation Initiative (OAEI). The results presented in this paper were obtained using benchmark ontologies used in OAEI conference. The results and analysis shows the significance of linguistic tools in the overall alignment process. The analysis presented in this paper can also be used as future guidelines for the research community who utilizes the linguistic matchers for finding the semantically equal entities from different ontologies.

Background and objective: Despite fast evolution cycles in deep learning methodologies for medical imaging in radiotherapy, auto-segmentation solutions rarely run in clinics due to the lack of open-source frameworks feasible for processing DICOM RT Structure Sets. Besides this shortage, available open-source DICOM RT Structure Set converters rely exclusively on 2D reconstruction approaches leading to pixelated contours with potentially low acceptance by healthcare professionals. PyRaDiSe, an open-source, deep learning framework independent Python package, addresses these issues by providing a framework for building auto-segmentation solutions feasible to operate directly on DICOM data. In addition, PyRaDiSe provides profound DICOM RT Structure Set conversion and processing capabilities; thus, it applies also to auto-segmentation-related tasks, such as dataset construction for deep learning model training.Methods: The PyRaDiSe package follows a holistic approach and provides DICOM data handling, deep learning model inference, pre-processing, and post-processing functionalities. The DICOM data handling allows for highly automated and flexible handling of DICOM image series, DICOM RT Structure Sets, and DICOM registrations, including 2D-based and 3D-based conversion from and to DICOM RT Structure Sets. For deep learning model inference, extending given skeleton classes is straightforwardly achieved, allowing for employing any deep learning framework. Furthermore, a profound set of pre-processing and post-processing routines is included that incorporate partial invertibility for restoring spatial properties, such as image origin or orientation.Results: The PyRaDiSe package, characterized by its flexibility and automated routines, allows for fast deployment and prototyping, reducing efforts for auto-segmentation pipeline implementation. Furthermore, while deep learning model inference is independent of the deep learning framework, it can easily be integrated into famous deep learning frameworks such as PyTorch or Tensorflow. The developed package has successfully demonstrated its capabilities in a research project at our institution for organs-at-risk segmentation in brain tumor patients. Furthermore, PyRaDiSe has shown its conversion performance for dataset construction.Conclusions: The PyRaDiSe package closes the gap between data science and clinical radiotherapy by enabling deep learning segmentation models to be easily transferred into clinical research practice. PyRaDiSe is available on https://github.com/ubern-mia/pyradise and can be installed directly from the Python Package Index using pip install pyradise.

Objective: To design a deep transfer learning framework for modeling fluence map predictions for stereotactic body radiation therapy (SBRT) of adrenal cancer and similar sites that usually have a small number of cases. Approach: We developed a transfer learning framework for adrenal SBRT planning that leverages knowledge in a pancreas SBRT planning model. Treatment plans from the two sites had different dose prescriptions and beam settings but both prioritized gastrointestinal sparing. A base framework was first trained with 100 pancreas cases. This framework consists of two convolutional neural networks (CNN), which predict individual beam doses (BD-CNN) and fluence maps (FM-CNN) sequentially for 9-beam intensity-modulated radiation therapy (IMRT) plans. Forty-five adrenal plans were split into training/validation/test sets with the ratio of 20/10/15. The base BD-CNN was re-trained with transfer learning using 5/10/15/20 adrenal training cases to produce multiple candidate adrenal BD-CNN models. The base FM-CNN was directly used for adrenal cases. The deep learning (DL) plans were evaluated by several clinically relevant dosimetric endpoints, producing a percentage score relative to the clinical plans. Main results: Transfer learning significantly reduced the number of training cases and training time needed to train such a DL framework. The adrenal transfer learning model trained with 5/10/15/20 cases achieved validation plan scores of 85.4/91.2/90.7/89.4, suggesting that model performance saturated with 10 training cases. Meanwhile, a model using all 20 adrenal training cases without transfer learning only scored 80.5. For the final test set, the 5/10/15/20-case models achieved scores of 73.5/75.3/78.9/83.3. Significance: It is feasible to use deep transfer learning to train an IMRT fluence prediction framework. This technique could adapt to different dose prescriptions and beam configurations. This framework potentially enables DL modeling for clinical sites that have a limited dataset, either due to few cases or due to rapid technology evolution.

Since different ontologies are mostly developed independently, establishing meaningful links between their entities, so-called ontology matching, is critical to ensure their communication. Due to the complexity of the ontology matching problem, Evolutionary Algorithm (EA) can present a good methodology for determining ontology alignments. However, since none of the similarity measure can distinguish all the identical ontology entities in any context, ontology alignments generated by the automatic matching tools should be validated by the users to ensure their qualities. To improve the quality of the ontology alignment, in this work, a similarity measure is first proposed to calculate the similarity value of two ontology entities; then, an optimal model for ontology matching problem is constructed; after that, an EA-based automatic matcher is presented to solve the ontology matching problem, which can also adaptively determine the timing of getting user involved; and finally, the ontology concept hierarchy graph based reasoning approaches are proposed to tradeoff the user's workload and his work's effect. The experiment is conducted on the Interactive track provided by the Ontology Alignment Evaluation Initiative (OAEI), and the comparisons with OAEI's participants show the effectiveness of our proposal.

Deep learning (DL) frameworks serve as the backbone for a wide range of artificial intelligence applications. However, bugs within DL frameworks can cascade into critical issues in higher-level applications, jeopardizing reliability and security. While numerous techniques have been proposed to detect bugs in DL frameworks, research exploring common API patterns across frameworks and the potential risks they entail remains limited. Notably, many DL frameworks expose similar APIs with overlapping input parameters and functionalities, rendering them vulnerable to shared bugs, where a flaw in one API may extend to analogous APIs in other frameworks. To address this challenge, we propose MirrorFuzz, an automated API fuzzing solution to discover shared bugs in DL frameworks. MirrorFuzz operates in three stages: First, MirrorFuzz collects historical bug data for each API within a DL framework to identify potentially buggy APIs. Second, it matches each buggy API in a specific framework with similar APIs within and across other DL frameworks. Third, it employs large language models (LLMs) to synthesize code for the API under test, leveraging the historical bug data of similar APIs to trigger analogous bugs across APIs. We implement MirrorFuzz and evaluate it on four popular DL frameworks (TensorFlow, PyTorch, OneFlow, and Jittor). Extensive evaluation demonstrates that MirrorFuzz improves code coverage by 39.92% and 98.20% compared to state-of-the-art methods on TensorFlow and PyTorch, respectively. Moreover, MirrorFuzz discovers 315 bugs, 262 of which are newly found, and 80 bugs are fixed, with 52 of these bugs assigned CNVD IDs.

A comprehensive empirical study on bug characteristics of deep learning frameworks

Antimicrobial peptides (AMPs) are vital components of innate immunotherapy. Existing approaches mainly rely on either deep learning for the automatic extraction of sequence features or traditional manual amino acid features combined with machine learning. The peptide sequence contains symmetrical sequence motifs or repetitive amino acid patterns, which may be related to the function and structure of the peptide. Recently, the advent of large language models has significantly boosted the representational power of sequence pattern features. In light of this, we present a novel AMP predictor called UniproLcad, which integrates three prominent protein language models—ESM-2, ProtBert, and UniRep—to obtain a more comprehensive representation of protein features. UniproLcad utilizes deep learning networks, encompassing the bidirectional long and short memory network (Bi-LSTM) and one-dimensional convolutional neural networks (1D-CNN), while also integrating an attention mechanism to enhance its capabilities. These deep learning frameworks, coupled with pre-trained language models, efficiently extract multi-view features from antimicrobial peptide sequences and assign attention weights to them. Through ten-fold cross-validation and independent testing, UniproLcad demonstrates competitive performance in the field of antimicrobial peptide identification. This integration of diverse language models and deep learning architectures enhances the accuracy and reliability of predicting antimicrobial peptides, contributing to the advancement of computational methods in this field.

Accurate prediction and confirmation of mechanical properties are critical for the reliable design and deployment of sustainable aluminum–agro-waste composites (AWCs). This study presents a confirmatory analytical–experimental–deep machine learning (DML) framework for the identification of true material properties under multi-response conditions. Experimental characterization yielded mean tensile strength of 190.61 MPa, compressive strength of 287.16 MPa, flexural strength of 243.94 MPa, wear rate of 5.46 × 10⁻⁶ mm³ /N·m, coefficient of friction of 0.34, and fatigue life of 5.86 × 10⁵ cycles, with low variability observed in strength-related properties and higher dispersion in tribological and fatigue responses. Isolated analytical predictions produced comparable estimates for tensile (189.60 MPa), compressive (296.42 MPa), flexural (255.75 MPa), and fatigue properties (5.97 × 10⁵ cycles), but failed to reliably capture tribological behavior, yielding a non-physical negative wear rate (− 4.06 mm³/N·m) and an overestimated coefficient of friction (0.45). Consolidated analytical modeling improved agreement for tensile (187.60 MPa), compressive (285.63 MPa), flexural (244.88 MPa), and fatigue life (5.86 × 10⁵ cycles), yet significantly deviated in tribological predictions, producing unrealistically high wear rate (0.20 mm³/N·m) and coefficient of friction (2.84). To resolve these discrepancies, DML models—implemented via a deep convolutional neural network (ID-CNN) and a hybrid residual CNN guided by a residual-based loss function—were employed to reconcile analytical (isolated/consolidated) predictions with experimental data. The DML framework successfully minimized analytical–experimental residuals and confirmed true predictive reliability for compressive strength, flexural strength, coefficient of friction, and fatigue life, while consistently identifying wear rate as falsely predicted across analytical frameworks. These findings establish DML as both a predictive and diagnostic tool, enabling discrimination between truly confirmed and analytically unstable properties. The proposed framework offers a scalable, data-informed pathway for optimized composite design, process refinement, and industrial adoption, with potential extension to multi-scale modeling and real-time process monitoring. Residual-based DML loss quantifies and minimizes analytical–experimental prediction errors. • Residual-based DML loss quantifies and minimizes analytical–experimental prediction errors. • Integrated analytical–DML modeling confirms strength, fatigue, and friction properties. • Bulk properties converge, while wear rate shows persistent tribological uncertainty. • The framework supports reliable certification and sustainable composite design.

As a key representation model of knowledge, ontology has been widely used in a lot of NLP related tasks, such as semantic parsing, information extraction and text mining etc. In this paper, we study the task of ontology matching, which concentrates on finding semantically related entities between different ontologies that describe the same domain, to solve the semantic heterogeneity problem. Previous works exploit different kinds of descriptions of an entity in ontology directly and separately to find the correspondences without considering the higher level correlations between the descriptions. Besides, the structural information of ontology haven’t been utilized adequately for ontology matching. We propose in this paper an ontology matching approach, named ERSOM, which mainly includes an unsupervised representation learning method based on the deep neural networks to learn the general representation of the entities and an iterative similarity propagation method that takes advantage of more abundant structure information of the ontology to discover more mappings. The experimental results on the datasets from Ontology Alignment Evaluation Initiative (OAEI1) show that ERSOM achieves a competitive performance compared to the state-of-the-art ontology matching systems. The OAEI is an international initiative organizing annual campaigns for evaluating ontology matching systems. All of the ontologies provided by OAEI are described in OWL-DL language, and like most of the other participates our ERSOM also manages the OWL ontology in its current version. OAEI: http://oaei.ontologymatching.org/

Ontology matching is usually performed to find semantic correspondences between the entity elements of different ontologies to enable interoperability. Current research on ontology matching has largely focused on representation learning. However, there still exist two limitations. Firstly, they are only used in the element level matching phase, ignoring relations of the entity. Secondly, the final alignment threshold is usually determined manually within these methods. It is difficult for an expert to adjust the threshold value and even more for non-expert user. To address these issues, we propose an alternative ontology matching framework, which models the matching process by embedding techniques with jointly encoding ontology terminological description and network structure. We further improve our iterative final alignment method by introducing an automatic adjustment of threshold method. Finally, we perform an experimental evaluation and compare it with state-of-the-art ontology matching systems on four Ontology Alignment Evaluation Initiative (OAEI) datasets. Our approach performs better than most of the systems and achieves a competitive performance. Moreover, we obtained F-measure values of 93.8% and 90.8% on the OAEI Large Biomedical Ontologies FMA-NCI and FMA-SNOMED subtasks.

BackgroundOne of the longest running tracks in the Ontology Alignment Evaluation Initiative is the Anatomy track which focuses on aligning two anatomy ontologies. The Anatomy track was started in 2005. In 2005 and 2006 the task in this track was to align the Foundational Model of Anatomy and the OpenGalen Anatomy Model. Since 2007 the ontologies used in the track are the Adult Mouse Anatomy and a part of the NCI Thesaurus. Since 2015 the data in the Anatomy track is also used in the Interactive track of the Ontology Alignment Evaluation Initiative.ResultsIn this paper we focus on the Anatomy track in the years 2007–2016 and the Anatomy part of the Interactive track in 2015–2016. We describe the data set and the changes it went through during the years as well as the challenges it poses for ontology alignment systems. Further, we give an overview of all systems that participated in the track and the techniques they have used. We discuss the performance results of the systems and summarize the general trends.ConclusionsAbout 50 systems have participated in the Anatomy track. Many different techniques were used. The most popular matching techniques are string-based strategies and structure-based techniques. Many systems also use auxiliary information. The quality of the alignment has increased for the best performing systems since the beginning of the track and more and more systems check the coherence of the proposed alignment and implement a repair strategy. Further, interacting with an oracle is beneficial.

Ontology matching is an effective technique to solve the ontology heterogeneous problem in Semantic Web. Since different ontology matchers do not necessarily find the same correct correspondences, usually several competing matchers are applied to the same pair of entities in order to increase evidence towards a potential match or mismatch. How to select, combine and tune various ontology matchers to obtain the high quality ontology alignment is one of the main challenges in ontology matching domain. Recently, Evolutionary Algorithms (EA) has become the most suitable methodology to face this challenge, however, the huge memory consumption, slow convergence and premature convergence limit its application and reduce the solution’s quality. To this end, in this paper, we propose a Hybrid Population-based Incremental Learning algorithm (HPBIL) to automatically select, combine and tune different ontology matchers, which can overcome three drawbacks of EA based ontology matching techniques and improve the ontology alignment’s quality. In one hand, HPBIL makes use of a probabilistic representation of the population to perform the optimization process, which can significantly reduce EA’s the memory consumption and the possibility of the premature convergence. In the other, we introduce the local search strategy into PBIL’s evolving process to trade off its exploration and exploitation, and this marriage between global search and local search is helpful to reduce the runtime. In the experiment, we utilize different scale testing cases provided by the Ontology Alignment Evaluation Initiative (OAEI 2016) to test HPBIL’s performance, and the experimental results show that HPBIL’s results significantly outperform other EA based ontology matching techniques and top-performers of the OAEI competitions.

With the explosive growth in generating data in the hydrographical domain, many hydrography ontologies have been developed and maintained to describe hydrographical features and the relationships between them. However, the existing hydrography ontologies are developed with varying project perspectives and objectives, which inevitably results in the differences in terms of knowledge representation. Determining various relationships between two entities in different ontologies offers the opportunity to link hydrographical data for multiple purposes, though the research on this topic is in its infancy. Different from the traditional ontology alignment whose cardinality is 1:1, i.e. one source ontology entity is mapping with one target ontology entity and vice versa, and the relationship is the equivalence, matching hydrography ontologies is a more complex task, whose cardinality could be 1:1, 1:n or m:n and the relationships could be equivalence or subsumption. To efficiently optimize the ontology alignment, in this paper, a discrete optimal model is first constructed for the ontology matching problem, and then a Compact Particle Swarm Optimization algorithm (CPSO) based matching technique is proposed to efficiently solve it. CPSO utilizes the compact real-value encoding and decoding mechanism and the objective-decomposing strategy to approximate the PSO’s evolving process, which can dramatically reduce PSO’s memory consumption and runtime while at the same time ensure the solution’s quality. The experiment exploits the Hydrography dataset in Complex track provided by the Ontology Alignment Evaluation Initiative (OAEI) to test our proposal’s performance. The experimental results show that CPSO-based approach can effectively reduce PSO’s runtime and memory consumption, and determine high-quality hydrography ontology alignments.

Ontology alignment is the task of identifying semantically equivalent entities from two given ontologies. Different ontologies have different representations of the same entity, resulting in a need to de-duplicate entities when merging ontologies. We propose a method for enriching entities in an ontology with external definition and context information, and use this additional information for ontology alignment. We develop a neural architecture capable of encoding the additional information when available, and show that the addition of external data results in an F1-score of 0.69 on the Ontology Alignment Evaluation Initiative (OAEI) largebio SNOMED-NCI subtask, comparable with the entity-level matchers in a SOTA system.

Nowadays, most real-world decision problems consist of two or more incommensurable or conflicting objectives to be optimized simultaneously, so-called multiobjective optimization problems (MOPs). Usually, a decision maker (DM) prefers only a single optimum solution in the Pareto front (PF), and the PF’s knee solution is logically the one if there are no user-specific or problem-specific preferences. In this context, the biomedical ontology matching problem in the Semantic Web (SW) domain is investigated, which can be of help to integrate the biomedical knowledge and facilitate the translational discoveries. Since biomedical ontologies often own large-scale concepts with rich semantic meanings, it is difficult to find a perfect alignment that could meet all DM’s requirements, and usually, the matching process needs to trade-off two conflict objectives, i.e., the alignment’s recall and precision. To this end, in this work, the biomedical ontology matching problem is first defined as a MOP, and then a compact multiobjective particle swarm optimization algorithm driven by knee solution (CMPSO-K) is proposed to address it. In particular, a compact evolutionary mechanism is proposed to efficiently optimize the alignment’s quality, and a max-min approach is used to determine the PF’s knee solution. In the experiment, three biomedical tracks provided by Ontology Alignment Evaluation Initiative (OAEI) are used to test CMPSO-K’s performance. The comparisons with OAEI’s participants and PSO-based matching technique show that CMPSO-K is both effective and efficient.