Am 20.11.2018 wurde im Lichthof der Leibniz Universität Hannover Forschung und Startups rund um das Thema künstliche Intelligenz vorgestellt.  Zusammen mit dem Forschungszentrum L3S von der Universität Hannover und 1KLANG®, die Marke der PROJEKTIONISTEN® GmbH, das Forschungsprojekt Data4UrbanMobility vorstellen.
Interessierte wurden über die Ergebnisse aus dem Projekt in Bezug auf die entwickelten Apps und das Dashboard informiert (s. unter Ergebnisse: https://data4urbanmobility.l3s.uni-hannover.de/index.php/de/ergebnisse/).

Wir freuen uns über die spannenden Gespräche, die geführt werden konnten und die neuen Tester, die uns bei der Citizen Science App „MiC – Move in the City“ unterstützen möchten.

• ETCH: Generalizing Body Fitting to Clothed Humans via Equivariant Tightness. Li, Boqian; Feng, Haiwen; Cai, Zeyu; Black, Michael J.; Xiu, Yuliang (2025).
• Evaluation of potentially toxic elements (PTEs) contamination in seawater, sediment, and sea snails (Nerita articulata and Cerithidea obtusa) from Kukup Fishing Village, Johor, Malaysia. Tan, Ruo Han; Ng, Chuck Chuan; Gu, Tianze; Tek, Peggy Pei Yee (2025). 197(5) 565.

  Molluscs, being highly susceptible to potentially toxic elements (PTEs) and easily accessible for human consumption, play a critical role in research on PTE pollution. This study focuses on Kukup Fishing Village in Johor, Malaysia, to investigate the levels of Cd, Cu, and Pb in seawater, sediment, and the soft tissues and shells of the gastropods Nerita articulata (lined nerite snail) and Cerithidea obtusa (obtuse horn shell). All collected samples were chemically digested before being analysed using the flame atomic absorption spectrometer (F-AAS). It is found that the PTEs are mostly accumulated in soft tissues and shells, followed by sediments and seawater, respectively. Besides, the results revealed that all study locations exhibited moderate to severe pollution, and the PTE concentrations in sea snails exceeded the thresholds set by the Malaysian Food Regulations (1985). The mean concentrations of the studied elements in the seawater, soft tissues, and shells of two studied species were found to be in the order of Pbthinspace>thinspaceCuthinspace>thinspaceCd. Additionally, Pb was heavily accumulated in the soft tissues of both species: N. articulata (712.587--723.242 ppm) and C. obtusa (705.935--708.626 ppm). Compared to N. articulata, C. obtusa showed a higher capacity to accumulate Cd (3.702--4.350 ppm) and Cu (92.687--157.445 ppm), particularly in the soft tissues. The shell of N. articulata and the soft tissue of C. obtusa were identified as potential biomonitoring indicators for Cd and Cu, respectively. It is recommended that the local Malaysian authorities could strengthen environmental management, implement regular monitoring, and raise public awareness to minimise seafood consumption from polluted areas as these measures could aid to reduce pollution, protect marine ecosystems, and safeguard public health.
• Enhancement of Quality of Service in Underwater Wireless Sensor Networks. S, Vinayprasad M; N, Jayaram M (N. Meghanathan, ed.) (2025). 17(2) 53–70.

  Underwater Wireless sensor network (UWSN) has become a main topic in the research of underwater communication with more research challenges. One of the main issues in the UWSN communication process is Quality of Service (QoS). Therefore, for enhancing the QoS in the UWSN a novel Clustering Hello routing based Honey Badger GoogleNet (CHbHBG) model is proposed. Primarily, the required sensor hubs are placed in the underwater communication environment. Further, the energy usage of each node is monitored and energy-efficient cluster head is selected by the proposed mechanism. Moreover, the data rate resources were predicted and allocated at the channel using the fitness process of the model. The optimal allocation process improves the QoS in the network. To prove the efficacy of the system, the metrics including throughput, network lifetime, latency, energy consumption, PDR, transmission loss, and path creation time are validated and compared with the recent models. The developed model attained the higher network performance as 99.72% PDR, 949.2kbps throughput, 4004.31s network lifetime, and 230.84J energy consumption.
• Ontolearn---A Framework for Large-scale OWL Class Expression Learning in Python. Demir, Caglar; Baci, Alkid; Kouagou, N’Dah Jean; Sieger, Leonie Nora; Heindorf, Stefan; Bin, Simon; Blübaum, Lukas; Bigerl, Alexander; Ngonga Ngomo, Axel-Cyrille (2025).
• MARVEL-40M+: Multi-Level Visual Elaboration for High-Fidelity Text-to-3D Content Creation. Sinha, Sankalp; Khan, Mohammad Sadil; Usama, Muhammad; Sam, Shino; Stricker, Didier; Ali, Sk Aziz; Afzal, Muhammad Zeshan (2025).
• Visualizing Intracellular Localization of Natural-Product-Based Chemical Probes Using Click-Correlative Light and Electron Microscopy. Schaller, Eva; Hofmann, Julian; Maher, Pamela; Stigloher, Christian; Decker, Michael (2025). 20(3) 721–730.
• Bringing Diversity from Diffusion Models to Semantic-Guided Face Asset Generation. Cai, Yunxuan; Xiang, Sitao; Li, Zongjian; Chen, Haiwei; Zhao, Yajie (2025).
• SceneFactory: A Workflow-centric and Unified Framework for Incremental Scene Modeling. Yuan, Yijun; Bleier, Michael; Nüchter, Andreas (2025). 1–20.

  We present SceneFactory, a workflow-centric and unified framework for incremental scene modeling, that conveniently supports a wide range of applications, such as (unposed and/or uncalibrated) multi-view depth estimation, LiDAR completion, (dense) RGB-D/RGB-L/Mono/Depth-only reconstruction and SLAM. The workflow-centric design uses multiple blocks as the basis for constructing different production lines. The supported applications, i.e., productions avoid redundancy in their designs. Thus, the focus is placed on each block itself for independent expansion. To support all input combinations, our implementation consists of four building blocks that form SceneFactory: (1) tracking, (2) flexion, (3) depth estimation, and (4) scene reconstruction. The tracking block is based on Mono SLAM and is extended to support RGB-D and RGB-LiDAR (RGB-L) inputs. Flexion is used to convert the depth image (untrackable) into a trackable image. For general-purpose depth estimation, we propose an unposed & uncalibrated multi-view depth estimation model (U2 -MVD) to estimate dense geometry. U2-MVD exploits dense bundle adjustment to solve for poses, intrinsics, and inverse depth. A semantic-aware ScaleCov step is then introduced to complete the multi-view depth. Relying on U2-MVD, SceneFactory both supports user-friendly 3D creation (with just images) and bridges the applications of Dense RGB-D and Dense Mono. For high-quality surface and color reconstruction, we propose Dual-purpose Multi-resolutional Neural Points (DM-NPs) for the first surface accessible Surface Color Field design, where we introduce Improved Point Rasterization (IPR) for point cloud based surface query. We implement and experiment with SceneFactory to demonstrate its broad applicability and high flexibility. Its quality also competes or exceeds the tightly-coupled state of the art approaches in all tasks. We contribute the code to the community
• Forcing the world into machines. Geuter, Jürgen (2025).

  For generative AI this purpose is quite clear: It’s about putting pressure on human labour. If you can generate somewhat passable prose or images or software code using an “AI” you can either try to run your business with fewer people (who tend to want to be paid) or (more realistically) you can push people’s wages down by always pointing at “the AI” when someone wants a raise or does not want to do unpaid overtime. The actual problem generative AI is trying to solve is having to pay people for their work.
• Bridging the Communication Gap: Evaluating AI Labeling Practices for Trustworthy AI Development Fischer, Raphael; Wischnewski, Magdalena; van der Staay, Alexander; Poitz, Katharina; Janiesch, Christian; Liebig, Thomas (2025).
• App-basierte Lieferdienste in Deutschland: Warum Menschen Gig-Work aufnehmen und meist schnell wieder beenden. Friedrich, Martin; Helm, Ines; Jost, Ramona; Lang, Julia; Müller, Christoph (2025). 2025–04.

  App-basierte Lieferdienste haben sich in den letzten Jahren rasant ausgebreitet. Das hat auch die öffentliche Diskussion um schlechte Arbeitsbedingungen der dort beschäftigten Gig-Worker angefacht. Allerdings gibt es bisher wenige gesicherte Erkenntnisse darüber, was Menschen zur Aufnahme von Gig-Jobs bewegt. Über die Gründe zur Beendigung dieser meist kurzen Jobs ist ebenfalls wenig bekannt. Das IAB bringt mit Ergebnissen einer neuen Befragung Licht in dieses Dunkel.
• Towards a phenomenological ontology of war. Gilks, Mark (2025). First View 1–19.

  This paper offers a critique of war from an existentialist-phenomenological perspective. Drawing on Martin Heidegger’s theory of ontology and Maurice Merleau-Ponty’s theory of perception, it develops a framework which integrates war and the body – and thus ontology and embodiment – in Critical War Studies. Two arguments are advanced: first, that war is in so far as we embody it (implying that the way in which we embody it determines the way in which it is); second, that the embodiment of war is essentially an agential activity. Thereby, this paper provides impetus for an ontological and moral re-avowal of war in critical academic discourse (for understanding war not primarily as a tragic fate but as our shameful doing). This, in turn, facilitates new perspective for interpretation and critique – to the extent, for example, that understanding the logic of war’s agential embodiment discloses what would constitute, and be necessary for, its disembodiment. Moreover, the paper points to clear possibilities for future research – for clarifying, for instance, the ontological upheaval latent in the prospect of future war.
• Drivers of late Quaternary lake level fluctuations of Khyargas Nuur, western Mongolia - glacial meltwater discharge or atmospheric moisture supply?. Wolf, Dennis; Lehmkuhl, Frank; Schaubert, Viktor; Rahimzadeh, Neda; Frechen, Manfred; Stauch, Georg; Batkhishig, Ochirbat; Wegmann, Karl (2025). 359 109373.

  This study presents the first comprehensive late Quaternary chronology of lake level variations of the Khyargas Nuur (western Mongolia), based on a geomorphological approach supported by luminescence dating of relict shorelines and lacustrine sediments. The endorheic Khyargas Nuur in the Basin of Great Lakes is the ultimate sink of a sequential water and sediment cascade from the adjacent Mongolian Altai and Khangai Mountains. Several intercalated lakes repeatedly joined as one major paleolake, as evidenced by various morphological shoreline features. Situated in the mid-latitude Westerlies-dominated climate regime of Arid Central Asia and affected by the distant effects of the East Asian Summer Monsoon, the dynamic climate of the Basin of Great Lakes is determined by the co-evolution of these atmospheric circulation systems. Our observations comprise 11 distinct paleolake levels between +7 m and +188 m above modern lake level (aml). Calculations of paleolake extent and water volume emphasize the periodically enhanced inflow and gradual capture and abandonment of upstream-located lakes. In the regional to global paleoclimatic context, our results reveal three distinct phases of lake level dynamics: (i) A transgression to a maximum level of +129 m aml during Marine Isotope Stage 5c (104.7 ± 14.4–88.8 ± 12.7 ka) primarily controlled by increased atmospheric moisture supply from the Westerlies. (ii) A post-Last Glacial Maximum lake expansion to a level of +118 m aml around 14 ka, ultimately controlled by enhanced glacial meltwater discharge into the basin. This is followed by a lake regression throughout the late Glacial to early Holocene transition in response to a gradually decreasing meltwater supply and a drier climate. (iii) A late Holocene transgression to +15 m aml reflecting a general Holocene wetting trend across arid Central Asia, followed by small-scale level fluctuations post 2.6 ka.
• Utilising coastal blue carbon (CBC) to mitigate the climate crisis: Current status and future analysis of China. Gu, Tianze; Ng, Chuck Chuan (2025). 266 107699.
• Der abwesende Partner in der Sozialpartnerschaft: Veränderungen der Mitbestimmungspraxis in Betrieben mit chinesischen Investoren. Technical Report (84), Bian, Shuwen (2025).

  Wenngleich die Beteiligung chinesischer Unternehmen an ausländischen Unternehmen in den letzten Jahren deutlich abgenommen hat, bleiben die chinesischen Investitionsaktivitäten in Deutschland auf einem stabil hohen Niveau. Der Einstieg chinesischer Investoren hat tiefgreifende Veränderungen in der Unternehmenskultur deutscher Standorte zur Folge. Mit den Übernahmen entsenden die meisten chinesischen Gesellschafter chinesische Manager*innen in die Geschäftsführung der Tochtergesellschaften. Diese personellen Veränderungen in der lokalen Führungsebene haben direkte Auswirkungen auf die Betriebsabläufe und die betriebliche Mitbestimmung. Zwar gibt es keine Hinweise darauf, dass chinesische Gesellschafter und das Management die Arbeit der Mitbestimmungsakteure proaktiv verhindern, allerdings suchen sie auch nicht die Zusammenarbeit mit den Arbeitnehmervertretungen. Diese Abwesenheit in der Sozialpartnerschaft wird besonders problematisch, wenn Krisensituationen eine gemeinsame Lösungsfindung erfordern. Von besonderer Bedeutung sind insbesondere vier Faktoren, die sich negativ auf die betriebliche Mitbestimmung auswirken können: eingeschränkter Informationszugang, sprachliche Barrieren, paternalistische Unternehmensführung und der Einfluss von Parteirichtlinien.
• In Türkiye, as in other countries, ethical committee. Chips (2025).

  In Türkiye, as in other countries, ethical committee
• While legal regulations provide a framework for these committees. Chips (2025).

  While legal regulations provide a framework for these committees
• This deficiency, coupled with variations in committee. Chips (2025).

  This deficiency, coupled with variations in committee
• frequently results in inconsistent evaluations of similar. Chips (2025).

  frequently results in inconsistent evaluations of similar
• LOLA - An Open-Source Massively Multilingual Large Language Model. Srivastava, Nikit; Kuchelev, Denis; Ngoli, Tatiana Moteu; Shetty, Kshitij; Röder, Michael; Zahera, Hamada M.; Moussallem, Diego; Ngomo, Axel-Cyrille Ngonga O. Rambow, L. Wanner, M. Apidianaki, H. Al-Khalifa, B. D. Eugenio, S. Schockaert (eds.) (2025). 6420–6446.
• Natural Language Generation. Reiter, Ehud (2025).
• ENHANCING PERSONALIZED RECIPE RECOMMENDATION THROUGH MULTI CLASS CLASSIFICATION. Neelam, Harish; Veerella, Koushik Sai (H. Neelam; K. S. Veerella, eds.) (2024). 14(5) 11.

  This paper intends to address the challenge of personalized recipe recommendation in the realm of diverse culinary preferences. The problem domain involves recipe recommendations, utilizing techniques such as association analysis and classification. Association analysis explores the relationships and connections between different ingredients to enhance the user experience. Meanwhile, the classification aspect involves categorizing recipes based on user-defined ingredients and preferences. A unique aspect of the paper is the consideration of recipes and ingredients belonging to multiple classes, recognizing the complexity of culinary combinations. This necessitates a sophisticated approach to classification and recommendation, ensuring the system accommodates the nature of recipe categorization. The paper seeks not only to recommend recipes but also to explore the process involved in achieving accurate and personalized recommendations.
• FaVEL: Fact Validation Ensemble Learning. Qudus, Umair; Tatkeu Pekarou, Franck Lionel; Morim da Silva, Ana Alexandra; Röder, Michael; Ngonga Ngomo, Axel-Cyrille in Lecture Notes in Computer Science, M. Alam, M. Rospocher, M. van Erp, L. Hollink, G. A. Gesese (eds.) (2024). (Vol. 15370) 209–225.
• UniQ-Gen: Unified Query Generation across Multiple Knowledge Graphs. Vollmers, Daniel; Srivastava, Nikit; Zahera, Hamada M.; Moussallem, Diego; Ngonga Ngomo, Axel-Cyrille (2024).
• Inference over Unseen Entities, Relations and Literals on Knowledge Graphs. Demir, Caglar; Kouagou, N; Sharma, Arnab; Ngonga Ngomo, Axel-Cyrille (2024). arXiv–2410.
• Enhancing Answers Verbalization using Large Language Models. Vollmers, Daniel; Sharma, Parth; Zahera, Hamada M.; Ngonga Ngomo, Axel-Cyrille (2024).
• Evaluating Negation with Multi-way Joins Accelerates Class Expression Learning. Karalis, Nikolaos; Bigerl, Alexander; Demir, Caglar; Heidrich, Liss; Ngomo, Axel-Cyrille Ngonga in Lecture Notes in Computer Science, A. Bifet, J. Davis, T. Krilavicius, M. Kull, E. Ntoutsi, I. Zliobaite (eds.) (2024). (Vol. 14946) 199–216.
• IndEL: Indonesian Entity Linking Benchmark Dataset for General and Specific Domains. Gusmita, Ria Hari; Faruq, Muhammad; Mousallem, Diego; Ngonga Ngomo, Axel-Cyrille (2024).
• Efficient Evaluation of Conjunctive Regular Path Queries Using Multi-way Joins. Karalis, Nikolaos; Bigerl, Alexander; Heidrich, Liss; Sherif, Mohamed Ahmed; Ngonga Ngomo, Axel-Cyrille (2024).
• Plekhg5 controls the unconventional secretion of Sod1 by presynaptic secretory autophagy. Hutchings, Amy-Jayne; Hambrecht, Bita; Veh, Alexander; Giridhar, Neha Jadhav; Zare, Abdolhossein; Angerer, Christina; Ohnesorge, Thorben; Schenke, Maren; Selvaraj, Bhuvaneish T.; Chandran, Siddharthan; Sterneckert, Jared; Petri, Susanne; Seeger, Bettina; Briese, Michael; Stigloher, Christian; Bischler, Thorsten; Hermann, Andreas; Damme, Markus; Sendtner, Michael; Lüningschrör, Patrick (2024). 15(1) 8622-.

  Increasing evidence suggests an essential function for autophagy in unconventional protein secretion (UPS). However, despite its relevance for the secretion of aggregate-prone proteins, the mechanisms of secretory autophagy in neurons have remained elusive. Here we show that the lower motoneuron disease-associated guanine exchange factor Plekhg5 drives the UPS of Sod1. Mechanistically, Sod1 is sequestered into autophagosomal carriers, which subsequently fuse with secretory lysosomal-related organelles (LROs). Exocytosis of LROs to release Sod1 into the extracellular milieu requires the activation of the small GTPase Rab26 by Plekhg5. Deletion of Plekhg5 in mice leads to the accumulation of Sod1 in LROs at swollen presynaptic sites. A reduced secretion of toxic ALS-linked SOD1G93A following deletion of Plekhg5 in SOD1G93A mice accelerated disease onset while prolonging survival due to an attenuated microglia activation. Using human iPSC-derived motoneurons we show that reduced levels of PLEKHG5 cause an impaired secretion of ALS-linked SOD1. Our findings highlight an unexpected pathophysiological mechanism that converges two motoneuron disease-associated proteins into a common pathway.
• Multi-Resolution Diffusion Models for Time Series Forecasting. Shen, Lifeng; Chen, Weiyu; Kwok, James (2024).
• An OWL Ontology for Linguistic Phenomena with Applications to Gallo-Italic Dialects in Sicily. Cantone, Domenico; Di Caro, Vincenzo Nicolò; Longo, Cristiano; Menza, Salvatore; Asmundo, Marianna Nicolosi; Santamaria, Daniele Francesco in CEUR Workshop Proceedings, A. Bikakis, R. Ferrario, S. Jean, B. Markhoff, A. Mosca, M. N. Asmundo (eds.) (2024). (Vol. 3809)
• Top Manufacturer of Orthocoir Sheets, Coir Cake & China Coir. Coir, We (W. Coir, ed.) (2024).
• PCFWebUI: Data-driven WebUI for holistic decarbonization based on PCF-Tracking. Kumar, Ajay; Naumann, Marius; Henne, Kevin; Sherif, Mohamed Ahmed in CEUR Workshop Proceedings, D. Garijo, A. L. Gentile, A. Kurteva, A. Mannocci, F. Osborne, S. Vahdati (eds.) (2024). (Vol. 3759)
• Generating SPARQL from Natural Language Using Chain-of-Thoughts Prompting. Zahera, Hamada M.; Ali, Manzoor; Sherif, Mohamed Ahmed; Moussallem, Diego; Ngomo, Axel-Cyrille Ngonga in Studies on the Semantic Web, A. A. Salatino, M. Alam, F. Ongenae, S. Vahdati, A. L. Gentile, T. Pellegrini, S. Jiang (eds.) (2024). (Vol. 60) 353–368.
• Analysis of Common Supervised Learning Algorithms Through Application. Narula, Palak (2023).
• COBALT: A Content-Based Similarity Approach for Link Discovery over Geospatial Knowledge Graphs. Becker, Alexander; Ahmed, Abdullah; Sherif, Mohamed Ahmed; Ngonga Ngomo, Axel-Cyrille (2023).
• Parameterized Complexity of Weighted Team Definability. Kontinen, Juha; Mahmood, Yasir; Meier, Arne; Vollmer, Heribert (2023).
• Using Pre-trained Language Models for Abstractive DBPEDIA Summarization: A Comparative Study. Zahera, Hamada M.; Vitiugin, Fedor; Sherif, Mohamed Ahmed; Castillo, Carlos; Ngonga Ngomo, Axel-Cyrille (2023).
• Glacier thinning, recession and advance, and the associated evolution of a glacial lake between 1966 and 2021 at<scp>Austerdalsbreen</scp>, western<scp>Norway</scp>. Seier, Gernot; Abermann, Jakob; Andreassen, Liss M.; Carrivick, Jonathan L.; Kielland, Pål H.; Löffler, Karina; Nesje, Atle; Robson, Benjamin A.; Røthe, Torgeir O.; Scheiber, Thomas; Winkler, Stefan; Yde, Jacob C. (2023). 35(1) 394–414.

  The Jostedalsbreen is the largest ice cap in Norway and mainland Europe. Rapid retreat of many of its outlet glaciers since the 2000s has led to the formation of several glacial lakes. Processes causing the formation and expansion of glacial lakes and their interaction with a glacier and terminal moraine have not been widely addressed yet. In this study, we investigate the degradation of the front of the southeast-facing outlet glacier Austerdalsbreen. Based on a variety of remotely sensed data (UAV-based and airborne orthophotos and DEMs, satellite images), we analyze the coincident glacial and proglacial changes of Austerdalsbreen and quantify the evolution of this transition zone during the last decades. In particular, we focus on the short-term evolution of the glacial lake since 2010, we examine the context of a glacier advance in the 1990s, and we report long-term changes by utilizing 1960s imagery. We discuss the evolution and conditions of Austerdalsbreen compared to other outlet glaciers of Jostedalsbreen. Overall, the glacier terminus has experienced a recession in the last decades. The 1990s terminus advance was more restricted than at other nearby outlet glaciers due to glacier surface debris cover, which is a critical factor for the glacier and lake evolution. However, in the most recent period, since 2012, a distinct expansion of a glacial lake is quantifiable. Since the rates of glacier surface lowering also considerably increased since approximately 2017 and the glacier retreated since the beginning of the 2000s with a clear maximum length decrease in 2015, we interpret the recently formed glacial lake as a contributory factor of glacial changes.
• Glacier thinning, recession and advance, and the associated evolution of a glacial lake between 1966 and 2021 at<scp>Austerdalsbreen</scp>, western<scp>Norway</scp>. Seier, Gernot; Abermann, Jakob; Andreassen, Liss M.; Carrivick, Jonathan L.; Kielland, Pål H.; Löffler, Karina; Nesje, Atle; Robson, Benjamin A.; Røthe, Torgeir O.; Scheiber, Thomas; Winkler, Stefan; Yde, Jacob C. (2023). 35(1) 394–414.

  The Jostedalsbreen is the largest ice cap in Norway and mainland Europe. Rapid retreat of many of its outlet glaciers since the 2000s has led to the formation of several glacial lakes. Processes causing the formation and expansion of glacial lakes and their interaction with a glacier and terminal moraine have not been widely addressed yet. In this study, we investigate the degradation of the front of the southeast-facing outlet glacier Austerdalsbreen. Based on a variety of remotely sensed data (UAV-based and airborne orthophotos and DEMs, satellite images), we analyze the coincident glacial and proglacial changes of Austerdalsbreen and quantify the evolution of this transition zone during the last decades. In particular, we focus on the short-term evolution of the glacial lake since 2010, we examine the context of a glacier advance in the 1990s, and we report long-term changes by utilizing 1960s imagery. We discuss the evolution and conditions of Austerdalsbreen compared to other outlet glaciers of Jostedalsbreen. Overall, the glacier terminus has experienced a recession in the last decades. The 1990s terminus advance was more restricted than at other nearby outlet glaciers due to glacier surface debris cover, which is a critical factor for the glacier and lake evolution. However, in the most recent period, since 2012, a distinct expansion of a glacial lake is quantifiable. Since the rates of glacier surface lowering also considerably increased since approximately 2017 and the glacier retreated since the beginning of the 2000s with a clear maximum length decrease in 2015, we interpret the recently formed glacial lake as a contributory factor of glacial changes.
• Similar vegetation‐geomorphic disturbance feedbacks shape unstable glacier forelands across mountain regions. Eichel, Jana; Draebing, Daniel; Winkler, Stefan; Meyer, Nele (2023). 14(2)

  Glacier forelands are among the most rapidly changing landscapes on Earth. Stable ground is rare as geomorphic processes move sediments across large areas of glacier forelands for decades to centuries following glacier retreat. Yet, most ecological studies sample exclusively on stable terrain to fulfill chronosequence criteria, thus missing potential feedbacks between geomorphic disturbances and vegetation colonization. By influencing vegetation and soil development, such vegetation-geomorphic disturbance feedbacks could be crucial to understand glacier foreland ecosystem development in a changing climate. We surveyed vegetation and environmental properties, including geomorphic disturbance intensities, in 105 plots located on both stable and unstable moraine terrain in two geomorphologically active glacier forelands in New Zealand and Switzerland. Our plot data showed that geomorphic disturbance intensities permanently changed from high/moderate to low/stable when vegetation reached cover values of around 40%. Around this cover value, species with response and effect traits adapted to geomorphic disturbances dominated. This suggests that such species can act as “biogeomorphic” ecosystem engineers that stabilize ground through positive feedback loops. Across floristic regions, biogeomorphic ecosystem engineer traits creating ground stabilization, such as mat growth and association with mycorrhiza, are remarkably similar. Nonmetric multidimensional scaling revealed a linked sequence of decreasing geomorphic disturbance intensities and changing species composition from pioneer to late successional species. We interpret this linked geomorphic disturbance-vegetation succession sequence as “biogeomorphic succession,” a common successional pathway in unstable river and coastal ecosystems across the world. Soil and vegetation development were related to this sequence and only advanced once biogeomorphic ecosystem engineer species covered 40%–45% of a plot, indicating a crucial role of biogeomorphic ecosystem engineer stabilization. Different topoclimatic conditions could explain variance in biogeomorphic succession timescales and ecosystem engineer root traits between the glacier forelands. As glacier foreland ground is widely unstable, we propose to consider glacier forelands as “biogeomorphic ecosystems” in which ecosystem structure and function are shaped by geomorphic disturbances and their feedbacks with adapted plant species, similar to rivers and coasts.
• Similar vegetation‐geomorphic disturbance feedbacks shape unstable glacier forelands across mountain regions. Eichel, Jana; Draebing, Daniel; Winkler, Stefan; Meyer, Nele (2023). 14(2)

  Glacier forelands are among the most rapidly changing landscapes on Earth. Stable ground is rare as geomorphic processes move sediments across large areas of glacier forelands for decades to centuries following glacier retreat. Yet, most ecological studies sample exclusively on stable terrain to fulfill chronosequence criteria, thus missing potential feedbacks between geomorphic disturbances and vegetation colonization. By influencing vegetation and soil development, such vegetation-geomorphic disturbance feedbacks could be crucial to understand glacier foreland ecosystem development in a changing climate. We surveyed vegetation and environmental properties, including geomorphic disturbance intensities, in 105 plots located on both stable and unstable moraine terrain in two geomorphologically active glacier forelands in New Zealand and Switzerland. Our plot data showed that geomorphic disturbance intensities permanently changed from high/moderate to low/stable when vegetation reached cover values of around 40%. Around this cover value, species with response and effect traits adapted to geomorphic disturbances dominated. This suggests that such species can act as “biogeomorphic” ecosystem engineers that stabilize ground through positive feedback loops. Across floristic regions, biogeomorphic ecosystem engineer traits creating ground stabilization, such as mat growth and association with mycorrhiza, are remarkably similar. Nonmetric multidimensional scaling revealed a linked sequence of decreasing geomorphic disturbance intensities and changing species composition from pioneer to late successional species. We interpret this linked geomorphic disturbance-vegetation succession sequence as “biogeomorphic succession,” a common successional pathway in unstable river and coastal ecosystems across the world. Soil and vegetation development were related to this sequence and only advanced once biogeomorphic ecosystem engineer species covered 40%–45% of a plot, indicating a crucial role of biogeomorphic ecosystem engineer stabilization. Different topoclimatic conditions could explain variance in biogeomorphic succession timescales and ecosystem engineer root traits between the glacier forelands. As glacier foreland ground is widely unstable, we propose to consider glacier forelands as “biogeomorphic ecosystems” in which ecosystem structure and function are shaped by geomorphic disturbances and their feedbacks with adapted plant species, similar to rivers and coasts.
• REDFM: a Filtered and Multilingual Relation Extraction Dataset. Huguet Cabot, ‪Pere-Lluís; Tedeschi, Simone; Ngonga Ngomo, Axel-Cyrille; Navigli, Roberto A. Rogers, J. Boyd-Graber, N. Okazaki (eds.) (2023). 4326–4343.

  Relation Extraction (RE) is a task that identifies relationships between entities in a text, enabling the acquisition of relational facts and bridging the gap between natural language and structured knowledge. However, current RE models often rely on small datasets with low coverage of relation types, particularly when working with languages other than English.In this paper, we address the above issue and provide two new resources that enable the training and evaluation of multilingual RE systems. First, we present SRED\($^\textrmFM$\), an automatically annotated dataset covering 18 languages, 400 relation types, 13 entity types, totaling more than 40 million triplet instances. Second, we propose RED\($^\textrmFM$\), a smaller, human-revised dataset for seven languages that allows for the evaluation of multilingual RE systems. To demonstrate the utility of these novel datasets, we experiment with the first end-to-end multilingual RE model, mREBEL, that extracts triplets, including entity types, in multiple languages. We release our resources and model checkpoints at [\urlhttps://www.github.com/babelscape/rebel](\urlhttps://www.github.com/babelscape/rebel).
• Clifford Embeddings – A Generalized Approach for Embedding in Normed Algebras. Demir, Caglar; Ngomo, Axel-Cyrille Ngonga (2023).
• Native Execution of GraphQL Queries over RDF Graphs Using Multi-way Joins. Karalis, Nikolaos; Bigerl, Alexander; Ngonga Ngomo, Axel-Cyrille (2023).
• Hardware-agnostic computation for large-scale knowledge graph embeddings. Demir, Caglar; Ngomo, Axel-Cyrille Ngonga (2022). 13
• Hashing the Hypertrie: Space- and Time-Efficient Indexing for SPARQL in Tensors. Bigerl, Alexander; Conrads, Lixi; Behning, Charlotte; Saleem, Muhammad; Ngonga Ngomo, Axel-Cyrille U. Sattler, A. Hogan, M. Keet, V. Presutti, J. P. A. Almeida, H. Takeda, P. Monnin, G. Pirrò, C. d’Amato (eds.) (2022). 57–73.

  Time-efficient solutions for querying RDF knowledge graphs depend on indexing structures with low response times to answer SPARQL queries rapidly. Hypertries---an indexing structure we recently developed for tensor-based triple stores---have achieved significant runtime improvements over several mainstream storage solutions for RDF knowledge graphs. However, the space footprint of this novel data structure is still often larger than that of many mainstream solutions. In this work, we detail means to reduce the memory footprint of hypertries and thereby further speed up query processing in hypertrie-based RDF storage solutions. Our approach relies on three strategies: (1) the elimination of duplicate nodes via hashing, (2) the compression of non-branching paths, and (3) the storage of single-entry leaf nodes in their parent nodes. We evaluate these strategies by comparing them with baseline hypertries as well as popular triple stores such as Virtuoso, Fuseki, GraphDB, Blazegraph and gStore. We rely on four datasets/benchmark generators in our evaluation: SWDF, DBpedia, WatDiv, and WikiData. Our results suggest that our modifications significantly reduce the memory footprint of hypertries by up to 70% while leading to a relative improvement of up to 39% with respect to average Queries per Second and up to 740% with respect to Query Mixes per Hour.
• Consistent signals of a warming climate in occupancy changes of three insect taxa over 40 years in central Europe. Engelhardt, Eva Katharina; Biber, Matthias F.; Dolek, Matthias; Fartmann, Thomas; Hochkirch, Axel; Leidinger, Jan; Löffler, Franz; Pinkert, Stefan; Poniatowski, Dominik; Voith, Johannes; Winterholler, Michael; Zeuss, Dirk; Bowler, Diana E.; Hof, Christian (2022). 28(13) 3998–4012.
• Schmidt-hammer exposure-age dating: a review of principles and practice. Matthews, John A.; Winkler, Stefan (2022). 230 104038.

  Schmidt-hammer exposure-age dating (SHD) is a numerical, calibrated-age dating technique based on the degree of weathering of rock surfaces. The Schmidt hammer measures the hardness or strength of a rock surface by means of R-values (rebound values), which are produced as the hammer impacts the surface. Age calibration is carried out with reference to at least two surfaces of known age (control surfaces). Mean R-values then yield numerical estimates of the time elapsed since exposure of rock surfaces to subaerial weathering. Under favourable circumstances with a sufficiently large number of impacts, SHD has an age resolution of 300–500 years for rock surfaces exposed during the Holocene and Late Pleistocene, and a current age range of ~25 ka. Accurate independent dating of control surfaces and careful selection of sites to reduce the effects of lithological variability are essential. R-value distributions are an important aid in the interpretation of SHD ages from active and relict, synchronous and diachronous, single and multiple-event surfaces. Important research applications have been made within the fields of Quaternary geochronology, palaeoenvironmental reconstruction, geomorphology and subaerial processes, and landscape dynamics. SHD has been found particularly useful in relation to boulder deposits and bedrock surfaces in glacial and periglacial environments. There is much untapped potential in coastal, temperate and tropical landscapes wherever rock surfaces are exposed to undisturbed rock weathering. The advantages and limitations of SHD compare favourably with the complementary exposure-age technique of terrestrial cosmogenic nuclide dating (TCND). The capability to provide a large number of exposure ages at low cost, and non-destructively in sensitive natural and artefactual settings, are special strengths of SHD.
• Schmidt-hammer exposure-age dating: a review of principles and practice. Matthews, John A.; Winkler, Stefan (2022). 230 104038.

  Schmidt-hammer exposure-age dating (SHD) is a numerical, calibrated-age dating technique based on the degree of weathering of rock surfaces. The Schmidt hammer measures the hardness or strength of a rock surface by means of R-values (rebound values), which are produced as the hammer impacts the surface. Age calibration is carried out with reference to at least two surfaces of known age (control surfaces). Mean R-values then yield numerical estimates of the time elapsed since exposure of rock surfaces to subaerial weathering. Under favourable circumstances with a sufficiently large number of impacts, SHD has an age resolution of 300–500 years for rock surfaces exposed during the Holocene and Late Pleistocene, and a current age range of ~25 ka. Accurate independent dating of control surfaces and careful selection of sites to reduce the effects of lithological variability are essential. R-value distributions are an important aid in the interpretation of SHD ages from active and relict, synchronous and diachronous, single and multiple-event surfaces. Important research applications have been made within the fields of Quaternary geochronology, palaeoenvironmental reconstruction, geomorphology and subaerial processes, and landscape dynamics. SHD has been found particularly useful in relation to boulder deposits and bedrock surfaces in glacial and periglacial environments. There is much untapped potential in coastal, temperate and tropical landscapes wherever rock surfaces are exposed to undisturbed rock weathering. The advantages and limitations of SHD compare favourably with the complementary exposure-age technique of terrestrial cosmogenic nuclide dating (TCND). The capability to provide a large number of exposure ages at low cost, and non-destructively in sensitive natural and artefactual settings, are special strengths of SHD.
• Age, origin and palaeoclimatic implications of peri- and paraglacial boulder-dominated landforms in Rondane, South Norway. Marr, Philipp; Winkler, Stefan; Dahl, Svein Olaf; Löffler, Jörg (2022). 408 108251.

  Boulder-dominated landforms of periglacial, paraglacial and related origin constitute a valuable, but often unexplored source of palaeoclimatic and morphodynamic information. The timing of landform development initiation and its subsequent stabilization can be linked to past climatic conditions offering the potential to reconstruct cold climatic periods. In this study, Schmidt-hammer exposure-age dating (SHD) was applied to a variety of boulder-dominated landforms (sorted stripes, blockfield, rock-slope failure, paraglacial alluvial fan) in Rondane, eastern South Norway for the first time. On the basis of old and young control points a regional SHD calibration curve was established and successively utilized for the calculation of surface exposure ages for individual landforms. The chronological investigation of development and stabilization of the respective landforms permitted an assessment of Holocene climate variability in Rondane and its impact on overall landform evolution. Our obtained SHD age estimates ranged from 11.44 ± 1.22 ka (ST-D2) to 4.09 ± 1.51 ka (AF1) showing their inactive and relict character. Most surface exposure ages for sorted stripes clustered between 9.88 ± 1.35 ka and 9.25 ± 1.21 ka, hence indicating stabilization during the late stage of the Erdalen Event or shortly thereafter. It is inferred that the blockfield formed prior to the Last Glacial Maximum, was protected by cold-based ice throughout glaciation and shortly reactivated during the Erdalen Event only to subsequently becoming inactive. The surface exposure age of a rock-slope failure (7.58 ± 0.73 ka) falls into the early phase of the Holocene Thermal Maximum (HTM, ~8.0–5.0 ka). This indicates permafrost degradation and/or increasing hydrological pressure negatively influencing slope stability. The paraglacial alluvial fan with its four subsites yielded ages between 8.73 ± 1.63 ka and 4.09 ± 1.51 ka. The old exposure ages point to fan aggradation following regional deglaciation due to paraglacial processes, whereas the younger ages can be explained by increasing precipitation during the onset neoglaciation at ~4.0 ka. Our results underline the importance of meltwater for the activation of periglacial landforms in a continental climate and indicate that the Erdalen Event and immediately following onset of the HTM had major impact on landscape evolution in Rondane. Our obtained surface exposure ages from boulder-dominated landforms in Rondane give important insights into the local palaeoclimatic variability during the Holocene.
• Age, origin and palaeoclimatic implications of peri- and paraglacial boulder-dominated landforms in Rondane, South Norway. Marr, Philipp; Winkler, Stefan; Dahl, Svein Olaf; Löffler, Jörg (2022). 408 108251.

  Boulder-dominated landforms of periglacial, paraglacial and related origin constitute a valuable, but often unexplored source of palaeoclimatic and morphodynamic information. The timing of landform development initiation and its subsequent stabilization can be linked to past climatic conditions offering the potential to reconstruct cold climatic periods. In this study, Schmidt-hammer exposure-age dating (SHD) was applied to a variety of boulder-dominated landforms (sorted stripes, blockfield, rock-slope failure, paraglacial alluvial fan) in Rondane, eastern South Norway for the first time. On the basis of old and young control points a regional SHD calibration curve was established and successively utilized for the calculation of surface exposure ages for individual landforms. The chronological investigation of development and stabilization of the respective landforms permitted an assessment of Holocene climate variability in Rondane and its impact on overall landform evolution. Our obtained SHD age estimates ranged from 11.44 ± 1.22 ka (ST-D2) to 4.09 ± 1.51 ka (AF1) showing their inactive and relict character. Most surface exposure ages for sorted stripes clustered between 9.88 ± 1.35 ka and 9.25 ± 1.21 ka, hence indicating stabilization during the late stage of the Erdalen Event or shortly thereafter. It is inferred that the blockfield formed prior to the Last Glacial Maximum, was protected by cold-based ice throughout glaciation and shortly reactivated during the Erdalen Event only to subsequently becoming inactive. The surface exposure age of a rock-slope failure (7.58 ± 0.73 ka) falls into the early phase of the Holocene Thermal Maximum (HTM, ~8.0–5.0 ka). This indicates permafrost degradation and/or increasing hydrological pressure negatively influencing slope stability. The paraglacial alluvial fan with its four subsites yielded ages between 8.73 ± 1.63 ka and 4.09 ± 1.51 ka. The old exposure ages point to fan aggradation following regional deglaciation due to paraglacial processes, whereas the younger ages can be explained by increasing precipitation during the onset neoglaciation at ~4.0 ka. Our results underline the importance of meltwater for the activation of periglacial landforms in a continental climate and indicate that the Erdalen Event and immediately following onset of the HTM had major impact on landscape evolution in Rondane. Our obtained surface exposure ages from boulder-dominated landforms in Rondane give important insights into the local palaeoclimatic variability during the Holocene.
• Towards holistic Entity Linking: Survey and directions. Oliveira, Italo Lopes; Fileto, Renato; Speck, René; Garcia, Lu’is Paulo F.; Moussallem, Diego; Lehmann, Jens (2021). 95
• Automatically generating instructions from tutorials for search and user navigation. Heindorf, Stefan; Lipka, Nedim (2021).
• New evidence for active talus-foot rock glaciers at Øyberget, southern Norway, and their development during the Holocene. Nesje, Atle; Matthews, John A; Linge, Henriette; Bredal, Marie; Wilson, Peter; Winkler, Stefan (2021). 31(11–12) 1786–1796.

  Synthetic aperture radar interferometry (InSAR) measurements demonstrate that lobate, blocky depositional landforms, located in southern Norway at an altitude of ~530 m above sea level, with an estimated mean annual air temperature of ~1.6°C, currently exhibit deformation attributed to viscous creep. Five years of InSAR measurements for six lobes demonstrate average surface velocities of 1.2–22.0 mm/year with maximum rates of 17.5–55.6 mm/year. New Schmidt-hammer exposure-age dating (SHD) of two proximal lobes reveals mid-Holocene ages (7.6 ± 1.3 ka and 6.0 ± 1.2 ka), which contrast with the early-Holocene SHD and 10Be ages obtained previously from distal lobes, and late-Holocene SHD ages presented here from two adjacent talus slopes (2.3 ± 1.0 ka and 2.4 ± 1.0 ka). Although passive transport of boulders on the surfaces of these small, slow-moving rock glaciers affected by compressive flow means that the exposure ages are close to minimum estimates of the time elapsed since lobe inception, disturbance of boulders on rock glaciers is a source of potentially serious underestimates of rock-glacier age. Rock-glacier development at Øyberget began shortly after local deglaciation around 10 ka before present and continued throughout the Holocene in response to microclimatic undercooling within the coarse blocky surface layer of the talus and rock-glacier lobes. We suggest this enhanced cooling lowers mean annual surface-layer temperature by at least ~3.6°C, which is needed at such a low altitude to sustain sporadic permafrost and avoid fast thawing as atmospheric temperatures rise. Our results point to circumstances where inferences about rock glaciers as indicators of regional climate should be interpreted with caution, and where they may be less useful in palaeoclimatic reconstruction than previously thought.
• LAUREN - Knowledge Graph Summarization for Question Answering. Jalota, Rricha; Vollmers, Daniel; Moussallem, Diego; Ngomo, Axel-Cyrille Ngonga (2021).

  Besides the challenge that a human can ask one question in many different ways, a key aspect in Question Answering approaches over Knowledge Graphs (KGQA) is to deal with the vast amount of information present in the knowledge graphs. Modern real-world knowledge graphs contain nearly millions of entities and relationships. Additionally, they are enriched with new facts every day. However, not all facts are relevant for answering particular questions, thus fostering several challenges to KGQA systems, which require interpretable and query-able data. One solution to filtering the extra data in knowledge graphs is to rely on graph summarization techniques. Graph-based summarization approaches aim to resize knowledge graphs to be more concise and precise by storing only relevant information. In this paper, we propose a framework named LAUREN that applies different summarization techniques on knowledge graphs to be used in KGQA systems. Our experiments show that LAUREN summarizes large knowledge graphs such as DBpedia by 2 million entities and its summarization still achieves the same performance on both question answering and linking tasks compared to the complete DBpedia.
• ASSET: A Semi-supervised Approach for Entity Typing in Knowledge Graphs. Zahera, Hamada M.; Heindorf, Stefan; Ngonga Ngomo, Axel-Cyrille (2021).

  Entity typing in knowledge graphs (KGs) aims to infer missing types of entities and might be considered one of the most significant tasks of knowledge graph construction since type information is highly relevant for querying, quality assurance, and KG applications. While supervised learning approaches for entity typing have been proposed, they require large amounts of (manually) labeled data, which can be expensive to obtain. In this paper, we propose a novel approach for KG entity typing that leverages semi-supervised learning from massive unlabeled data. Our approach follows a teacher-student paradigm that allows combining a small amount of labeled data with a large amount of unlabeled data to boost performance. We conduct several experiments on two benchmarking datasets (FB15k-ET and YAGO43k-ET). Our results demonstrate the effectiveness of our approach in improving entity typing in KGs. Given type information for only 1% of entities, our approach ASSET predicts missing types with a F1-score of 0.47 and 0.64 on the datasets FB15k-ET and YAGO43k-ET, respectively, outperforming supervised baselines.
• I-AID: Identifying Actionable Information from Disaster-related Tweets. Zahera, Hamada M.; Jalota, Rricha; Sherif, Mohamed Ahmed; Ngonga Ngomo, Axel-Cyrille (2021).

  Social media plays a significant role in disaster management by providing valuable data about affected people, donations and help requests. Recent studies highlight the need to filter information on social media into fine-grained content labels. However, identifying useful information from massive amounts of social media posts during a crisis is a challenging task. In this paper, we propose I-AID, a multimodel approach to automatically categorize tweets into multi-label information types and filter critical information from the enormous volume of social media data. I-AID incorporates three main components: i) a BERT- based encoder to capture the semantics of a tweet and represent as a low-dimensional vector, ii) a graph attention network (GAT) to apprehend correlations between tweets' words/entities and the corresponding information types, and iii) a Relation Network as a learnable distance metric to compute the similarity between tweets and their corresponding information types in a supervised way. We conducted several experiments on two real publicly-available datasets. Our results indicate that I-AID outperforms state-of- the-art approaches in terms of weighted average F1 score by +6% and +4% on the TREC-IS dataset and COVID-19 Tweets, respectively.
• New evidence for active talus-foot rock glaciers at Øyberget, southern Norway, and their development during the Holocene. Nesje, Atle; Matthews, John A; Linge, Henriette; Bredal, Marie; Wilson, Peter; Winkler, Stefan (2021). 31(11–12) 1786–1796.

  Synthetic aperture radar interferometry (InSAR) measurements demonstrate that lobate, blocky depositional landforms, located in southern Norway at an altitude of ~530 m above sea level, with an estimated mean annual air temperature of ~1.6°C, currently exhibit deformation attributed to viscous creep. Five years of InSAR measurements for six lobes demonstrate average surface velocities of 1.2–22.0 mm/year with maximum rates of 17.5–55.6 mm/year. New Schmidt-hammer exposure-age dating (SHD) of two proximal lobes reveals mid-Holocene ages (7.6 ± 1.3 ka and 6.0 ± 1.2 ka), which contrast with the early-Holocene SHD and 10Be ages obtained previously from distal lobes, and late-Holocene SHD ages presented here from two adjacent talus slopes (2.3 ± 1.0 ka and 2.4 ± 1.0 ka). Although passive transport of boulders on the surfaces of these small, slow-moving rock glaciers affected by compressive flow means that the exposure ages are close to minimum estimates of the time elapsed since lobe inception, disturbance of boulders on rock glaciers is a source of potentially serious underestimates of rock-glacier age. Rock-glacier development at Øyberget began shortly after local deglaciation around 10 ka before present and continued throughout the Holocene in response to microclimatic undercooling within the coarse blocky surface layer of the talus and rock-glacier lobes. We suggest this enhanced cooling lowers mean annual surface-layer temperature by at least ~3.6°C, which is needed at such a low altitude to sustain sporadic permafrost and avoid fast thawing as atmospheric temperatures rise. Our results point to circumstances where inferences about rock glaciers as indicators of regional climate should be interpreted with caution, and where they may be less useful in palaeoclimatic reconstruction than previously thought.
• GATES: Using Graph Attention Networks for Entity Summarization. Firmansyah, Asep Fajar; Moussallem, Diego; Ngomo, Axel-Cyrille Ngonga (2021).

  The sheer size of modern knowledge graphs has led to increased attention being paid to the entity summarization task. Given a knowledge graph T and an entity e found therein, solutions to entity summarization select a subset of the triples from T which summarize e's concise bound description. Presently, the best performing approaches rely on sequence-to-sequence models to generate entity summaries and use little to none of the structure information of T during the summarization process. We hypothesize that this structure information can be exploited to compute better summaries. To verify our hypothesis, we propose GATES, a new entity summarization approach that combines topological information and knowledge graph embeddings to encode triples. The topological information is encoded by means of a Graph Attention Network. Furthermore, ensemble learning is applied to boost the performance of triple scoring. We evaluate GATES on the DBpedia and LMDB datasets from ESBM (version 1.2), as well as on the FACES datasets. Our results show that GATES outperforms the state-of-the-art approaches on 4 of 6 configuration settings and reaches up to 0.574 F-measure. Pertaining to resulted summaries quality, GATES still underperforms the state of the arts as it obtains the highest score only on 1 of 6 configuration settings at 0.697 NDCG score. An open-source implementation of our approach and of the code necessary to rerun our experiments are available at https://github.com/dice-group/GATES.
• Designing Ritual Artifacts for Technology-Mediated Relationship Transitions. Klüber, Sara; Löffler, Diana; Hassenzahl, Marc; Nord, Ilona; Hurtienne, Jörn in TEI ’20 (2020). 349–361.

  Rituals are ubiquitous but not commonplace, help people to make sense of their life, and cultivate personal or social meaning. Although secularization and digitalization impact the occurrence of formal rituals, the need for marking life's transitions remains unchanged. New rituals emerge, such as marking relationship status by hanging love locks on bridges. Tangible technologies hold great potential for augmenting, changing, or enhancing ritual practices which often involve enactments and symbolic props. In this paper, we analyze individual stories of hanging love locks and derive six pointers for designing technology-mediated relationship transition rituals. We applied the pointers in the design of El Corazón, a tangible artifact for relationship transition rituals. The results of an evaluation with 20 sweethearts show that relationship rituals can be designed deliberately, that tangibles can shape ritual experiences and that technology-mediated rituals can provide people with new means of coping with relationship uncertainty.
• Participant’s View: Short-Term Evaluation of Realizing PD Ideals. Klüber, Sara; Maas, Franzisca; Hohm, Anna; Hurtienne, Jörn in PDC ’20 (2020). 138–142.

  Participatory Design (PD) has long been described as a way to democratize technology development by involving those affected by the outcomes. Besides a good fit of technology as the outcome, PD allows participants to ‘have a say’, and supports ‘mutual learning’ and ‘co-realization’. A diverse range of PD methods has been developed, but there is a shortage of empirical studies showing whether (and why) these methods help pursue PD ideals. We therefore report on a case study of a short-term evaluation approach that may exemplary close this gap, and that allowed us to gain more knowledge on how our method affected participants. Participants perceived moderate to high feelings of ‘having a say’, ‘mutual learning’, and ‘co-realization’. In future research, mutual learning may exemplarily be improved by introducing a changing peer-to-peer procedure to the method. The evaluation may further be advanced by taking expectations into account.
• Experience Matters: Design and Evaluation of an Anesthesia Support Tool Guided by User Experience Theory. Klüber, Sara; Maas, Franzisca; Schraudt, David; Hermann, Gina; Happel, Oliver; Grundgeiger, Tobias in DIS ’20 (2020). 1523–1535.

  Despite many advances, clinical decision support tools (DSTs) often suffer from implementation and acceptance problems in the actual clinical context. We suggest that considering psychological needs-based and embodied user experience theories in the design of DSTs could help to overcome these problems. To examine this idea, we iteratively developed a DST called Cassandra supporting anesthetic teams in crisis management, specifically focusing on psychological needs and fluent interaction with the social and physical environment. We preliminarily evaluated Cassandra in a medical simulation, requiring anesthetic teams to handle a crisis. Although not all features of Cassandra had the intended effect, the results indicated that interacting with Cassandra supported the fulfillment of the identified needs for autonomy, competence, and relatedness and was seamlessly integrated into existing diagnostic processes. Considering user experience theories for the design of DSTs seems a promising way to overcome implementation and acceptance problems and eventually improve patient safety.
• MindPeaks: Formative Evaluation Method of Mindfulness Meditation Apps. Reinhardt, Daniel; Baur, Cordula; Klüber, Sara; Hurtienne, Jörn in DIS’ 20 Companion (2020). 37–42.

  Today's busy lifestyle can make us feel overwhelmed. Mindfulness meditation supports our well-being by slowing things down and drawing our attention to the present moment. Accordingly, a wealth of guided mindfulness meditation apps have emerged in the past years. Despite the many apps and an increasing interest within the HCI literature, there is little consensus of how to evaluate these apps and how to derive opportunities for re-design from the results. Our new MindPeaks method utilizes electroencephalography (EEG) to link design elements to meditative states and the user`s self-regulation of attention. This information can help designers to understand and optimize the efficacy of, for example, applications supporting mindfulness meditation.
• Finally on Par?! Multimodal and Unimodal Interaction for Open Creative Design Tasks in Virtual Reality. Zimmerer, Chris; Wolf, Erik; Wolf, Sara; Fischbach, Martin; Lugrin, Jean-Luc; Latoschik, Marc Erich (2020). 222–231.

  Multimodal Interfaces (MMIs) have been considered to provide promising interaction paradigms for Virtual Reality (VR) for some time. However, they are still far less common than unimodal interfaces (UMIs). This paper presents a summative user study comparing an MMI to a typical UMI for a design task in VR. We developed an application targeting creative 3D object manipulations, i.e., creating 3D objects and modifying typical object properties such as color or size. The associated open user task is based on the Torrence Tests of Creative Thinking. We compared a synergistic multimodal interface using speech-accompanied pointing/grabbing gestures with a more typical unimodal interface using a hierarchical radial menu to trigger actions on selected objects. Independent judges rated the creativity of the resulting products using the Consensual Assessment Technique. Additionally, we measured the creativity-promoting factors flow, usability, and presence. Our results show that the MMI performs on par with the UMI in all measurements despite its limited flexibility and reliability. These promising results demonstrate the technological maturity of MMIs and their potential to extend traditional interaction techniques in VR efficiently.
• HOBBIT: A platform for benchmarking Big Linked Data. Röder, Michael; Kuchelev, Denis; Ngonga Ngomo, Axel-Cyrille (2020). 3(1) 15–35.
• Squirrel - Crawling RDF Knowledge Graphs on the Web. Röder, Michael; de Souza Jr, Geraldo; Ngonga Ngomo, Axel-Cyrille in Lecture Notes in Computer Science, J. Z. Pan, V. Tamma, C. d’Amato, K. Janowicz, B. Fu, A. Polleres, O. Seneviratne, L. Kagal (eds.) (2020). (Vol. 12507) 34–47.
• Tentris -- A Tensor-Based Triple Store. Bigerl, Alexander; Conrads, Lixi; Behning, Charlotte; Sherif, Mohamed Ahmed; Saleem, Muhammad; Ngonga Ngomo, Axel-Cyrille J. Z. Pan, V. Tamma, C. d’Amato, K. Janowicz, B. Fu, A. Polleres, O. Seneviratne, L. Kagal (eds.) (2020). 56–73.

  The number and size of RDF knowledge graphs grows continuously. Efficient storage solutions for these graphs are indispensable for their use in real applications. We present such a storage solution dubbed Tentris. Our solution represents RDF knowledge graphs as sparse order-3 tensors using a novel data structure, which we dub hypertrie. It then uses tensor algebra to carry out SPARQL queries by mapping SPARQL operations to Einstein summation. By being able to compute Einstein summations efficiently, Tentris outperforms the commercial and open-source RDF storage solutions evaluated in our experiments by at least 1.8 times with respect to the average number of queries it can serve per second on three datasets of up to 1 billion triples. Our code, evaluation setup, results, supplementary material and the datasets are provided at https://tentris.dice-research.org/iswc2020.
• OPTIC: A Deep Neural Network Approach for Entity Linking using Word and Knowledge Embeddings. Oliveira, Italo Lopes; Moussallem, Diego; Garcia, Lu’is Paulo Faina; Fileto, Renato J. Filipe, M. Smialek, A. Brodsky, S. Hammoudi (eds.) (2020). 315–326.
• NABU - Multilingual Graph-Based Neural RDF Verbalizer. Moussallem, Diego; Gnaneshwar, Dwaraknath; Ferreira, Thiago Castro; Ngomo, Axel-Cyrille Ngonga in Lecture Notes in Computer Science, J. Z. Pan, V. Tamma, C. d’Amato, K. Janowicz, B. Fu, A. Polleres, O. Seneviratne, L. Kagal (eds.) (2020). (Vol. 12506) 420–437.
• Under the Greenwood Tree Hardy, Thomas in The Cambridge Edition of the Novels and Stories of Thomas Hardy, (S. Gatrell, ed.) (2019). Cambridge University Press.
• Finding Datasets in Publications: The University of Paderborn Approach. Jalota, Rricha; Srivastava, Nikit; Vollmers, Daniel; Speck, René; Röder, Michael; Usbeck, Ricardo; Ngonga Ngomo, Axel-Cyrille J. I. Lane, I. Mulvany, P. Nathan (eds.) (2019).
• Authorization Framework for Medical Data. Madadevaiah, Geetha (2019). 11(3) 7–28.

  In this paper, the authors describe an approach for sharing sensitive medical data with the consent of the data owner. The framework builds on the advantages of the Semantic Web technologies and makes it secure and robust for sharing sensitive information in a controlled environment. The framework uses a combination of Role-Based and Rule-Based Access Policies to provide security to a medical data repository as per the FAIR guidelines. A lightweight ontologywas developed, to collect consent from the users indicating which part of their data they want to share with another user having a particular role. Here, the authors have considered the scenario of sharing the medical data by the owner of data, say the patient, with relevant persons such as physicians, researchers, pharmacist, etc. To prove this concept, the authors developed a prototype and validated using the Sesame OpenRDF Workbench with 202,908 triples and a consent graph stating consents per patient.
• Developing a Framework for Online Practice Examination and Automated Score Generation. Sani, S. M. Saniul Islam; Karim, Rezaul; Arefin, Mohammad Shamsul (2018). abs/0806.3765
• Joint proceedings of the 4th Workshop on Semantic Deep Learning (SemDeep-4) and NLIWoD4: Natural Language Interfaces for the Web of Data (NLIWOD-4) and 9th Question Answering over Linked Data challenge (QALD-9) co-located with 17th International Semantic Web Conference (ISWC 2018), Monterey, California, United States of America, October 8th - 9th, 2018 Choi, Key-Sun; Anke, Luis Espinosa; Declerck, Thierry; Gromann, Dagmar; Kim, Jin-Dong; Ngonga Ngomo, Axel-Cyrille; Saleem, Muhammad; Usbeck, Ricardo in CEUR Workshop Proceedings (2018). (Vol. 2241) CEUR-WS.org.
• Proceedings of the 3rd International Workshop on Geospatial Linked Data and the 2nd Workshop on Querying the Web of Data co-located with 15th Extended Semantic Web Conference (ESWC 2018), Heraklion, Greece, June 3, 2018 Wauer, Matthias; Sherif, Mohamed Ahmed; Saleem, Muhammad; Hartig, Olaf; Usbeck, Ricardo; Verborgh, Ruben; Ngomo, Axel-Cyrille Ngonga in CEUR Workshop Proceedings (2018). (Vol. 2110) CEUR-WS.org.
• Towards a Semantic Message-driven Microservice Platform for Geospatial and Sensor Data. Wauer, Matthias; Sherif, Mohamed Ahmed; Ngonga Ngomo, Axel-Cyrille in CEUR Workshop Proceedings, M. Wauer, M. A. Sherif, M. Saleem, O. Hartig, R. Usbeck, R. Verborgh, A.-C. N. Ngomo (eds.) (2018). (Vol. 2110) 47–58.
• The Tale of Sansa Spark. Ermilov, Ivan; Lehmann, Jens; Sejdiu, Gezim; Bühmann, Lorenz; Westphal, Patrick; Stadler, Claus; Bin, Simon; Chakraborty, Nilesh; Petzka, Henning; Saleem, Muhammad; Ngonga Ngomo, Axel-Cyrille; Jabeen, Hajira (2017).
• Using Multi-Label Classification for Improved Question Answering. Usbeck, Ricardo; Hoffmann, Michael; Röder, Michael; Lehmann, Jens; Ngomo, Axel-Cyrille Ngonga (2017).
• Second RDF Stream Processing and the Querying the Web of Data Workshops Calbimonte, JP; Tran, MD; Dell’Aglio, D; Le Phuoc, D; Saleem, Muhammad; Usbeck, Ricardo; Verborgh, Ruben; Ngonga Ngomo, Axel-Cyrille (2017).
• Joint Proceedings of BLINK2017: 2nd International Workshop on Benchmarking Linked Data and NLIWoD3: Natural Language Interfaces for the Web of Data co-located with 16th International Semantic Web Conference (ISWC 2017), Vienna, Austria, October 21st - to - 22nd, 2017 Usbeck, Ricardo; Ngomo, Axel-Cyrille Ngonga; Kim, Jin-Dong; Choi, Key-Sun; Cimiano, Philipp; Fundulaki, Irini; Krithara, Anastasia in CEUR Workshop Proceedings (2017). (Vol. 1932) CEUR-WS.org.
• Self-Wiring Question Answering Systems. Usbeck, Ricardo; Huthmann, Jonathan; Duldhardt, Nico; Ngonga Ngomo, Axel-Cyrille (2016).
• Managing Geospatial Linked Data in the GeoKnow Project. Lehmann, Jens; Athanasiou, Spiros; Both, Andreas; Garcia-Rojas, Alejandra; Giannopoulos, Giorgos; Hladky, Daniel; Hoeffner, Konrad; Grange, Jon Jay Le; Ngomo, Axel-Cyrille Ngonga; Sherif, Mohamed Ahmed; Stadler, Claus; Wauer, Matthias; Westphal, Patrick; Zaslawski, Vadim T. Narock, P. Fox (eds.) (2015). (Vol. 20) 51–78.
• Using Caching for Local Link Discovery on Large Data Sets. Hassan, Mofeed M.; Speck, René; Ngonga Ngomo, Axel-Cyrille P. Cimiano, F. Frasincar, G.-J. Houben, D. Schwabe (eds.) (2015). (Vol. 9114) 344–354.
• MANAGING WEB SERVICES COMMUNITIES: A CACHE FOR QUERIES OPTIMISATION. Limam, Hela; Akaichi, Jalel (2014). 1(1) 01–19.

  With the advance of Web Services technologies and the mergence of Web Services into the information space, tremendous opportunities for empowering users and organizations appear in various application domains including electronic commerce, travel, intelligence information gathering and analysis, health care, digital government, etc. However, the technology to organize, search, integrate these Web Services has not kept pace with the rapid growth of the available information space. The number of Web Services to be integrated may be large and continuously changing. The ubiquitous need for Web Services integration across heterogeneous information sources pushed Web Services providers to join each others into Web Services Communities. Different approaches related to the specification, management and querying of a community were proposed. Whoever, current approaches for managing and querying Web Services Communities’ lack of a general formal and clear design. Hence, we propose a framewok which enables the community management then allows community members to share reusable fragments of kow-how in order to optimize query processing among communities .To demonstrate the viability of our approach, we propose a health care community infrastructure for supporting a community management and querying as well as a prototype application that utilizes communities.
• The Design and Implementation of Probabilistic Programming Languages. Goodman, Noah D.; Stuhlmüller, Andreas (2014).

  Probabilistic programming languages (PPLs) unify techniques for the formal description of computation and for the representation and use of uncertain knowledge. PPLs have seen recent interest from the artificial intelligence, programming languages, cognitive science, and natural languages communities. This book explains how to implement PPLs by lightweight embedding into a host language. We illustrate this by designing and implementing WebPPL, a small PPL embedded in Javascript. We show how to implement several algorithms for universal probabilistic inference, including priority-based enumeration with caching, particle filtering, and Markov chain Monte Carlo. We use program transformations to expose the information required by these algorithms, including continuations and stack addresses. We illustrate these ideas with examples drawn from semantic parsing, natural language pragmatics, and procedural graphics.
• AGDISTIS - Agnostic Disambiguation of Named Entities Using Linked Open Data. Usbeck, Ricardo; Ngonga Ngomo, Axel-Cyrille; Röder, Michael; Gerber, Daniel; Coelho, Sandro Atha’ide; Auer, Sören; Both, Andreas in Frontiers in Artificial Intelligence and Applications, T. Schaub, G. Friedrich, B. O’Sullivan (eds.) (2014). (Vol. 263) 1113–1114.
• From RDF to Natural Language and Back. Gerber, Daniel; Ngonga Ngomo, Axel-Cyrille P. Buitelaar, P. Cimiano (eds.) (2014). 193–209.
• Evaluating topic coherence measures. Rosner, Frank; Hinneburg, Alexander; Röder, Michael; Nettling, Martin; Both, Andreas (2014).
• Semantic Quran: A Multilingual Resource for Natural-Language Processing. Sherif, Mohamed Ahmed; Ngonga Ngomo, Axel-Cyrille (2014). 6(4) 339–345.

  In this paper we describe the Semantic Quran dataset, a multilingual RDF representation of translations of the Quran. The dataset was created by integrating data from two different semi-structured sources and aligned to an ontology designed to represent multilingual data from sources with a hierarchical structure. The resulting RDF data encompasses 43 different languages which belong to the most under-represented languages in the Linked Data Cloud, including Arabic, Amharic and Amazigh. We designed the dataset to be easily usable in natural-language processing applications with the goal of facilitating the development of knowledge extraction tools for these languages. In particular, the Semantic Quran is compatible with the Natural-Language Interchange Format and contains explicit morpho-syntactic information on the utilized terms. We present the ontology devised for structuring the data. We also provide the transformation rules implemented in our extraction framework. Finally, we detail the link creation process as well as possible usage scenarios for the Semantic Quran dataset.
• Ensemble Learning for Named Entity Recognition. Speck, René; Ngomo, Axel-Cyrille Ngonga in Lecture Notes in Computer Science, P. Mika, T. Tudorache, A. Bernstein, C. Welty, C. A. Knoblock, D. Vrandecic, P. Groth, N. F. Noy, K. Janowicz, C. A. Goble (eds.) (2014). (Vol. 8796) 519–534.
• The Principles and Practice of Probabilistic Programming. Goodman, Noah D. (2013). 399–402.
• Publishing and Interlinking the Global Health Observatory Dataset. Zaveri, Amrapali; Lehmann, Jens; Auer, Sören; Hassan, Mofeed M.; Sherif, Mohamed Ahmed; Martin, Michael (2013). (3) 315–322.

  The improvement of public health is one of the main indicators for societal progress. Statistical data for monitoring public health is highly relevant for a number of sectors, such as research (e.g. in the life sciences or economy), policy making, health care, pharmaceutical industry, insurances etc. Such data is meanwhile available even on a global scale, e.g. in the Global Health Observatory (GHO) of the United Nations's World Health Organization (WHO). GHO comprises more than 50 different datasets, it covers all 198 WHO member countries and is updated as more recent or revised data becomes available or when there are changes to the methodology being used. However, this data is only accessible via complex spreadsheets and, therefore, queries over the 50 different datasets as well as combinations with other datasets are very tedious and require a significant amount of manual work. By making the data available as RDF, we lower the barrier for data re-use and integration. In this article, we describe the conversion and publication process as well as use cases, which can be implemented using the GHO data.
• Evaluating the disparity between active areas of biomedical research and the global burden of disease employing Linked Data and data-driven discovery. Zaveri, Amrapali; Pietrobon, Ricardo; Ermilov, Timofey; Martin, Michael; Heino, Norman; Auer, Sören H.Herre, R.Hoehndorf, J.Kelso, S.Schulz (eds.) (2010).
• Enhanced Generic Information Services Using Mobile Messaging. Saleem, Muhammad; Zahir, Ali; Ismail, Yasir; Saeed, Bilal in Lecture Notes in Computer Science, P. Bellavista, R.-S. Chang, H.-C. Chao, S.-F. Lin, P. M. A. Sloot (eds.) (2010). (Vol. 6104) 510–521.
• Disentangling the Wikipedia Category Graph for Corpus Extraction. Ngonga Ngomo, Axel-Cyrille; Schumacher, Frank (2009). 39(2) 5–10.
• Disentangling the Wikipedia Category Graph for Corpus Extraction. Ngomo, Axel-Cyrille Ngonga; Schumacher, Frank (2009). 39 5–10.
• Towards an Implicit and Collaborative Evolution of Terminological Ontologies. Ngonga Ngomo, Axel-Cyrille E. Bolisani (ed.) (2008). 65–88.
• Adaptive and Context-Sensitive Information Retrieval. Ngonga Ngomo, Axel-Cyrille S. Hawamdeh (ed.) (2007). (Vol. 5) 289–300.