@techreport{66447, abstract = {{Volatility modeling is utilized across numerous fields including finance, environmental studies, and social sciences. It is particularly relevant in scenarios where understanding and predicting conditional variability is crucial, such as when dealing with incremental or time-dependent data. In this paper, novel short and long memory volatility models of the EGARCH family are introduced and analyzed, which are closely related to the well-established EGARCH model proposed by Nelson (1991) but share desirable theoretical properties in several dimensions. Recently developed members of the so-called EGARCH family, which introduces a modulus-log transformation proposed by John and Draper (1980) and a power transformation for the size and magnitude effect to tackle the problem with near-zero innovations and the asymmetric impact of positive and negative shocks on the volatility, are discussed. After a theoretical discussion of the proposed and related volatility models, the practical performance of the elaborated volatility models is compared to well-established and traditional GARCH approaches. A general QMLE algorithm is proposed to estimate the model parameters. The practical relevance of the advanced models is illustrated through a comparative study. By applying these volatility models to a variety of international stock index returns, this paper identifies market-specific characteristics as well as unique strengths and weaknesses of discussed volatility models. Although the practical performance of the recently introduced models is comparable to those obtained by the traditional EGARCH model, they generally outperform traditional non-exponential volatility models used as benchmarks and thus provide a useful alternative to existing short and long memory volatility models. }}, author = {{Hanke, Dominik Christian and Feng, Yuanhua and Uhde, André}}, keywords = {{Modulus Log-GARCH, Modified (FI)EGARCH, Modulus asymmetric (FI)Log-GARCH, (FI)EGARCH, long memory, modulus-log transformation, QMLE, model selection, implementation in R}}, title = {{{Comparing the behaviors of some original short and long memory exponential volatility models}}}, year = {{2026}}, } @inproceedings{62007, abstract = {{Ensemble methods are widely employed to improve generalization in machine learning. This has also prompted the adoption of ensemble learning for the knowledge graph embedding (KGE) models in performing link prediction. Typical approaches to this end train multiple models as part of the ensemble, and the diverse predictions are then averaged. However, this approach has some significant drawbacks. For instance, the computational overhead of training multiple models increases latency and memory overhead. In contrast, model merging approaches offer a promising alternative that does not require training multiple models. In this work, we introduce model merging, specifically weighted averaging, in KGE models. Herein, a running average of model parameters from a training epoch onward is maintained and used for predictions. To address this, we additionally propose an approach that selectively updates the running average of the ensemble model parameters only when the generalization performance improves on a validation dataset. We evaluate these two different weighted averaging approaches on link prediction tasks, comparing the state-of-the-art benchmark ensemble approach. Additionally, we evaluate the weighted averaging approach considering literal-augmented KGE models and multi-hop query answering tasks as well. The results demonstrate that the proposed weighted averaging approach consistently improves performance across diverse evaluation settings.}}, author = {{Sapkota, Rupesh and Demir, Caglar and Sharma, Arnab and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Proceedings of the Thirteenth International Conference on Knowledge Capture(K-CAP 2025)}}, keywords = {{Knowledge Graphs, Embeddings, Ensemble Learning}}, location = {{Dayton, OH, USA}}, publisher = {{ACM}}, title = {{{Parameter Averaging in Link Prediction}}}, doi = {{https://doi.org/10.1145/3731443.3771365}}, year = {{2025}}, } @inproceedings{60497, abstract = {{Despite the advantages that the virtual knowledge graph paradigm has brought to many application domains, state-of-the-art systems still do not support popular graph database management systems like Neo4j. Their query rewriting algorithms focus on languages like conjunctive queries and their unions, which were developed for relational data and are poorly suited for graph data. Moreover, they also limit the expressiveness of the ontology languages that admit rewritings, restricting them to those that enjoy the so-called FO-rewritability property. Rewritings have thus focused on the DL-Lite family of Description Logics. In this paper, we propose a technique for rewriting a family of navigational queries for a suitably tailored fragment of ELHI. Leveraging navigational features in the target query language, we can include some widely-used axiom shapes not supported by DL-Lite. We implemented a proof-of-concept prototype that rewrites into Cypher queries, and tested it on a real-world cognitive neuroscience use case with promising results.}}, author = {{Löhnert, Bianca and Augsten, Nikolaus and Okulmus, Cem and Ortiz, Magdalena}}, booktitle = {{The Semantic Web - 22nd European Semantic Web Conference, {ESWC} 2025, Portoroz, Slovenia, June 1-5, 2025, Proceedings, Part {I}}}, isbn = {{9783031945748}}, issn = {{0302-9743}}, keywords = {{Ontology-based Data Access, Property Graphs, Navigational Queries}}, location = {{Portorož, Slovenia}}, pages = {{342----361}}, publisher = {{Springer Nature Switzerland}}, title = {{{Towards Practicable Algorithms for Rewriting Graph Queries Beyond DL-Lite}}}, doi = {{10.1007/978-3-031-94575-5_19}}, volume = {{15718}}, year = {{2025}}, } @article{61123, abstract = {{Knowledge graphs are used by a growing number of applications to represent structured data. Hence, evaluating the veracity of assertions in knowledge graphs—dubbed fact checking—is currently a challenge of growing importance. However, manual fact checking is commonly impractical due to the sheer size of knowledge graphs. This paper is a systematic survey of recent works on automatic fact checking with a focus on knowledge graphs. We present recent fact-checking approaches, the varied sources they use as background knowledge, and the features they rely upon. Finally, we draw conclusions pertaining to possible future research directions in fact checking knowledge graphs.}}, author = {{Qudus, Umair and Röder, Michael and Saleem, Muhammad and Ngonga Ngomo, Axel-Cyrille}}, issn = {{0360-0300}}, journal = {{ACM Computing Surveys}}, keywords = {{fact checking, knowledge graphs, fact-checkers, check worthiness, evidence retrieval, trust, veracity.}}, publisher = {{Association for Computing Machinery (ACM)}}, title = {{{Fact Checking Knowledge Graphs -- A Survey}}}, doi = {{10.1145/3749838}}, volume = {{58}}, year = {{2025}}, } @inproceedings{50479, abstract = {{Verifying assertions is an essential part of creating and maintaining knowledge graphs. Most often, this task cannot be carried out manually due to the sheer size of modern knowledge graphs. Hence, automatic fact-checking approaches have been proposed over the last decade. These approaches aim to compute automatically whether a given assertion is correct or incorrect. However, most fact-checking approaches are binary classifiers that fail to consider the volatility of some assertions, i.e., the fact that such assertions are only valid at certain times or for specific time intervals. Moreover, the few approaches able to predict when an assertion was valid (i.e., time-point prediction approaches) rely on manual feature engineering. This paper presents TEMPORALFC, a temporal fact-checking approach that uses multiple sources of background knowledge to assess the veracity and temporal validity of a given assertion. We evaluate TEMPORALFC on two datasets and compare it to the state of the art in fact-checking and time-point prediction. Our results suggest that TEMPORALFC outperforms the state of the art on the fact-checking task by 0.13 to 0.15 in terms of Area Under the Receiver Operating Characteristic curve and on the time-point prediction task by 0.25 to 0.27 in terms of Mean Reciprocal Rank. Our code is open-source and can be found at https://github.com/dice-group/TemporalFC.}}, author = {{Qudus, Umair and Röder, Michael and Kirrane, Sabrina and Ngomo, Axel-Cyrille Ngonga}}, booktitle = {{The Semantic Web – ISWC 2023}}, editor = {{R. Payne, Terry and Presutti, Valentina and Qi, Guilin and Poveda-Villalón, María and Stoilos, Giorgos and Hollink, Laura and Kaoudi, Zoi and Cheng, Gong and Li, Juanzi}}, isbn = {{9783031472398}}, issn = {{0302-9743}}, keywords = {{temporal fact checking · ensemble learning · transfer learning · time-point prediction · temporal knowledge graphs}}, location = {{Athens, Greece}}, pages = {{465–483}}, publisher = {{Springer, Cham}}, title = {{{TemporalFC: A Temporal Fact Checking Approach over Knowledge Graphs}}}, doi = {{10.1007/978-3-031-47240-4_25}}, volume = {{14265}}, year = {{2023}}, } @article{23456, author = {{Mattiolo, Davide and Steffen, Eckhard}}, issn = {{0364-9024}}, journal = {{Journal of Graph Theory}}, keywords = {{factorization, perfect matchings, regular graphs, r-graphs}}, number = {{1}}, pages = {{107--116}}, title = {{{Highly edge‐connected regular graphs without large factorizable subgraphs}}}, doi = {{10.1002/jgt.22729}}, volume = {{99}}, year = {{2021}}, } @inproceedings{10586, abstract = {{We consider the problem of transforming a given graph G_s into a desired graph G_t by applying a minimum number of primitives from a particular set of local graph transformation primitives. These primitives are local in the sense that each node can apply them based on local knowledge and by affecting only its 1-neighborhood. Although the specific set of primitives we consider makes it possible to transform any (weakly) connected graph into any other (weakly) connected graph consisting of the same nodes, they cannot disconnect the graph or introduce new nodes into the graph, making them ideal in the context of supervised overlay network transformations. We prove that computing a minimum sequence of primitive applications (even centralized) for arbitrary G_s and G_t is NP-hard, which we conjecture to hold for any set of local graph transformation primitives satisfying the aforementioned properties. On the other hand, we show that this problem admits a polynomial time algorithm with a constant approximation ratio.}}, author = {{Scheideler, Christian and Setzer, Alexander}}, booktitle = {{Proceedings of the 46th International Colloquium on Automata, Languages, and Programming}}, keywords = {{Graphs transformations, NP-hardness, approximation algorithms}}, location = {{Patras, Greece}}, pages = {{150:1----150:14}}, publisher = {{Dagstuhl Publishing}}, title = {{{On the Complexity of Local Graph Transformations}}}, doi = {{10.4230/LIPICS.ICALP.2019.150}}, volume = {{132}}, year = {{2019}}, } @inproceedings{15921, abstract = {{Ranking plays a central role in a large number of applications driven by RDF knowledge graphs. Over the last years, many popular RDF knowledge graphs have grown so large that rankings for the facts they contain cannot be computed directly using the currently common 64-bit platforms. In this paper, we tackle two problems: Computing ranks on such large knowledge bases efficiently and incrementally. First, we present D-HARE, a distributed approach for computing ranks on very large knowledge graphs. D-HARE assumes the random surfer model and relies on data partitioning to compute matrix multiplications and transpositions on disk for matrices of arbitrary size. Moreover, the data partitioning underlying D-HARE allows the execution of most of its steps in parallel. As very large knowledge graphs are often updated periodically, we tackle the incremental computation of ranks on large knowledge bases as a second problem. We address this problem by presenting I-HARE, an approximation technique for calculating the overall ranking scores of a knowledge without the need to recalculate the ranking from scratch at each new revision. We evaluate our approaches by calculating ranks on the 3 × 10^9 and 2.4 × 10^9 triples from Wikidata resp. LinkedGeoData. Our evaluation demonstrates that D-HARE is the first holistic approach for computing ranks on very large RDF knowledge graphs. In addition, our incremental approach achieves a root mean squared error of less than 10E−7 in the best case. Both D-HARE and I-HARE are open-source and are available at: https://github.com/dice-group/incrementalHARE. }}, author = {{Desouki, Abdelmoneim Amer and Röder, Michael and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Proceedings of the 30th ACM Conference on Hypertext and Social Media - HT '19}}, isbn = {{9781450368858}}, keywords = {{Knowledge Graphs, Ranking, RDF}}, pages = {{163--171}}, publisher = {{ACM}}, title = {{{Ranking on Very Large Knowledge Graphs}}}, doi = {{10.1145/3342220.3343660}}, year = {{2019}}, } @article{48877, abstract = {{OpenML is an online machine learning platform where researchers can easily share data, machine learning tasks and experiments as well as organize them online to work and collaborate more efficiently. In this paper, we present an R package to interface with the OpenML platform and illustrate its usage in combination with the machine learning R package mlr (Bischl et al. J Mach Learn Res 17(170):1—5, 2016). We show how the OpenML package allows R users to easily search, download and upload data sets and machine learning tasks. Furthermore, we also show how to upload results of experiments, share them with others and download results from other users. Beyond ensuring reproducibility of results, the OpenML platform automates much of the drudge work, speeds up research, facilitates collaboration and increases the users’ visibility online.}}, author = {{Casalicchio, Giuseppe and Bossek, Jakob and Lang, Michel and Kirchhoff, Dominik and Kerschke, Pascal and Hofner, Benjamin and Seibold, Heidi and Vanschoren, Joaquin and Bischl, Bernd}}, issn = {{0943-4062}}, journal = {{Computational Statistics}}, keywords = {{Databases, Machine learning, R, Reproducible research}}, number = {{3}}, pages = {{977–991}}, title = {{{OpenML: An R Package to Connect to the Machine Learning Platform OpenML}}}, doi = {{10.1007/s00180-017-0742-2}}, volume = {{34}}, year = {{2019}}, } @inproceedings{39538, abstract = {{This article discusses the application of Pictorial Janus (PJ) for the rapid development and analysis of protocols by animation and complete visualization. In order to make PJ applicable in the context of hardware description we first extend PJ by timing facilities (Timed PJ) and introduce an approach for integrating VHDL models into this visual framework preserving the simulation semantics of VHDL. We finally give the example of the specification and animation of a non interlocked protocol.}}, author = {{Müller, Wolfgang and Lehrenfeld, Georg and Tahedl, C.}}, booktitle = {{Proceedings of ASP-DAC'95/CHDL'95/VLSI'95 with EDA Technofair}}, isbn = {{4-930813-67-0}}, keywords = {{Animation, Protocols, Timing, Computer languages, Electronic mail, Context modeling, Visualization, Control systems, Flow graphs, Trademarks}}, title = {{{Complete Visual Specification and Animations of Protocols}}}, doi = {{10.1109/ASPDAC.1995.486383}}, year = {{1995}}, }