5916 research outputs found
Sort by
Clustering-based negative sampling approaches for protein-protein interaction prediction
No full textThe lack of confirmed negative interactions poses a major challenge to the prediction of protein-protein interactions. The reliable selection of these negative samples within a dataset is crucial for a better understanding of the underlying patterns and dynamics. The random sampling method is the most widely used negative sampling method, where negative pairs are randomly selected from unlabelled samples (i.e., samples not experimentally confirmed as positive interactions). However, they tend to introduce inaccurately labelled negative samples, resulting in less reliable predictions, which may affect the efficiency of the learning process. Our study aims to assess the reliability of clustering-based negative sampling methods and highlight their fundamental differences from the widely used random sampling method. To achieve this goal, we propose a hierarchical clustering-based algorithm that uses different mechanisms to select negative instances from unlabelled instances. We investigated the effectiveness of our proposed approach compared to existing clustering-based negative sampling methods and random sampling on four different datasets. The results indicate that clustering-based methods surpass the commonly used random sampling method.TÜBİTA
Pagerank-based unsupervised deep vertex representations for anti-money laundering detection
No full textAnti-money laundering is an international web of laws, regulations, and procedures aimed at uncovering money that has been disguised as legitimate income. Strict anti-money laundering (AML) laws and procedures require major and continuous transaction observation in inferring possible illegal events. Nevertheless, traditional rule-based approaches in banks frequently generate a significant number of false positives, which impose a major burden. In this case, deep learning approaches, especially graph-based Graph Neural Network-based (GNN) methods, could be explored in generating better anti-money laundering results. Here, we propose a diffusion-based AMLPD, which is novel in generating unsupervised node embeddings via learning graph embeddings inductively while detecting AML. AMLPD assumes a direction between edges, and it incorporates vertex and edge feature knowledge while encoding graph's structure knowledge. AMLPD infers a vertex's local state via combining diffusion with PageRank, which is an important knowledge for AML when embedded into low dimensional space Then, our approach can detect AMLs by a classifier using this low dimensional representation. Our approach can be scaled to larger data, as well as it can help with explainable AI by facilitating the embeddings analysis. According to experiments, our approach outperforms the baseline approaches. Therefore, AMLPD is favourable in enhancing the quality of GNN-based AML identification
Participant perceptions of a robotic coach conducting positive psychology exercises: A qualitative analysis
No full textThis article presents a qualitative analysis of participants' perceptions of a robotic coach conducting Positive Psychology exercises, providing insights for the future design of robotic coaches. Participants (n = 20) took part in a single-session (avg. 31 +/- 10 minutes) Human-Robot Interaction study in a laboratory setting. We created the design of the robotic coach, and its affective adaptation, based on user-centred design research and collaboration with a professional coach. We transcribed post-study participant interviews and conducted a Thematic Analysis. We discuss the results of that analysis, presenting aspects participants found particularly helpful (e.g., the robot asked the correct questions and helped them think of new positive things in their life), and what should be improved (e.g., the robot's utterance content should be more responsive). We found that participants had no clear preference for affective adaptation or no affective adaptation, which may be due to both positive and negative user perceptions being heightened in the case of adaptation. Based on our qualitative analysis, we highlight insights for the future design of robotic coaches, and areas for future investigation (e.g., examining how participants with different personality traits, or participants experiencing isolation, could benefit from an interaction with a robotic coach).Publisher versio
Inferring effort-safety trade off in perturbed squat-to-stand task by reward parameter estimation
No full textIn this study, an inverse reinforcement learning (IRL) method is developed to estimate the parameters of a reward function that is assumed to guide the movement of a biological or artificial agent. The workings of the method is shown on the problem of estimating the effort-safety trade-off of humans during perturbed squat-to-stand motions based on their Center of Mass (COM) trajectories. The proposed method involves data generation by reinforcement learning (RL) and a novel data augmentation mechanism followed by neural network training. After the training, the neural network acts as the reward parameter estimator given the Center of Mass (COM) trajectories as input. The performance of the developed method is assessed through systematic simulation experiments, where it is shown that the parameter estimation made by our method is significantly more accurate than the baseline of an optimized template-based IRL approach. In addition, as a proof of concept, a set of human movement data is analyzed with the developed method. The results revealed that most participants acquired a strategy that ensures low effort expenditure with a safety margin, producing COM trajectories slightly away from the effort-optimal
Mean-field interacting systems with sequential coalescence at future ensemble averages
No full textWe introduce a new family of coalescent mean-field interacting particle systems by producing a pinning property that acts over a chosen sequence of multiple time segments. Throughout their evolution, these stochastic particles converge in time (i.e. get pinned) to their random ensemble average at the termination point of any one of the given time segments, only to burst back into life and repeat the underlying principle of convergence in each of the successive time segments, until they are fully exhausted. Although the architecture is represented by a system of piecewise stochastic differential equations, we prove that the conditions generating the pinning property enable every particle to preserve their continuity over their entire lifetime almost surely. As the number of particles in the system increases asymptotically, the system decouples into mutually independent diffusions, which, albeit displaying progressively uncorrelated behaviour, still close in on, and recouple at, a deterministic value at each termination point. Finally, we provide additional analytics including a universality statement for our framework, a study of what we call adjourned coalescent mean-field interacting particles, a set of results on commutativity of double limits, and a proposal of what we call covariance waves.Publisher versio
Seismic performance evaluation and retrofit of a liquid storage tank in high seismic region
No full textThis study investigates seismic performance of existing liquid pentane storage tanks located in a tank farm in the Kocaeli region of Turkey, known for its high seismic activity. The tanks are constructed to slide freely on the reinforced concrete foundation. The evaluation of the tanks’ seismic performance is carried out using three-dimensional (3D) finite element methods with nonlinear time-history analysis. The developed finite element model for assessing the seismic performance of the existing tanks considers dynamic interactions between the tank and its foundation, as well as the potential for tank supports to uplift and slide over the foundation. The study identifies significant deficiencies in the seismic performance of the existing tanks, primarily attributed to the lack of tank foundation anchorage. The tanks are retrofitted to improve tanks seismic behavior by anchoring the tanks supports to the reinforced concrete foundation using four Hilti anchors with injectable epoxy and prefabricated steel split sleeves that require no field welding. The retrofitted tanks undergo seismic evaluation to validate the effectiveness of the retrofit strategy and its impact on the tanks’ seismic behavior. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025
Excellence in hotel businesses: The case of a european quality award-winning hotel
No full textThe hotel industry is a useful domain to apply quality management systems given the structure of properties with numerous stakeholders and processes. The inclination of hotel businesses towards excellence models is low despite a high rate of usage of other quality management systems. This study therefore aims to showcase a European Quality Award-Winning hotel – AlpenResort Schwarz – and offer insights into the implementation of and results and benefits from the European Foundation for Quality Management Excellence Model (EFQM EM) to enable hotel businesses to become aware of the excellence concept and the model as a tool to improve their competitive advantage and sustainable performance. The major finding of the study is that the EFQM EM is effective for independent hotels, enabling small businesses to gain an international view when improving their managerial systems in accordance with sustainable development goals. Copyright © 2025 Inderscience Enterprises Ltd.Özyeğin Üniversitesi ; Fin Universit
Unlocking product complexity: Elevating emotional exhaustion and customer relationship performance through strategic ties
No full textDrawing upon the theoretical framework of the job demands-resources model, we extend and validate a conceptual model linking product complexity, interfirm ties, intrafirm ties, emotional exhaustion, and customer relationship performance. We conceptualize product complexity as a job demand and interfirm and intrafirm ties as personal job resources. We elucidate the mediating mechanism through which product complexity influences customer relationship performance via emotional exhaustion, as well as its boundary conditions. Utilizing a two-study approach (Study 1: 244 sales personnel, Study 2: 268 sales personnel), this research reveals that product complexity heightens salesperson emotional exhaustion. Furthermore, we offer empirical support indicating that intrafirm ties negatively moderate the relationship between product complexity and emotional exhaustion. Conversely, we demonstrate that interfirm ties mitigate the adverse impact of product complexity on emotional exhaustion. Lastly, our findings indicate that the indirect association between product complexity and customer relationship performance via emotional exhaustion is contingent upon both intrafirm and interfirm ties
Quantum security mechanisms for defense applications
No full textMany international organizations, such as the EU and NATO, and national governments have launched strategic initiatives supporting the migration to quantum-safe cryptography. Examples of such initiatives are Spanish CCN (Cybersecurity Defence of Spanish Cyberspace) endorsing post-quantum cryptographic algorithms like FrodoKEM and CRYSTALS-Kyber, and the European Union promoting secure quantum infrastructure through programs like Quantum Flagship, OpenQKD, and EuroQCI. Moreover, the efforts supported by the European Defence Fund reinforce the strategic importance of early quantum-resilient adoption across civil and defense sectors, through projects such as the Disruptive SDN secure communications for European Defence (DISCRETION) or the Quantum Agile and Resilient Military Communications (Q-ARM). This paper focuses particularly on defense-related quantum communication technologies, assessing not only quantum key distribution but also advanced primitives like quantum oblivious transfer, quantum digital signatures, and quantum secure direct communication, among others. It highlights their respective vulnerabilities, criticality, and potential countermeasures. Hybrid solutions that combine classical and quantum-resistant encryption methods and cryptoagility are recommended to enhance resilience
DA-Mamba: Domain ddaptive hybrid mamba-transformer based one-stage object detection
No full textRecent 2D CNN-based domain adaptation approaches struggle with long-range dependencies due to limited receptive fields, making it difficult to adapt to target domains with significant spatial distribution changes. While transformer-based domain adaptation methods better capture distant relationships through self-attention mechanisms that facilitate more effective cross-domain feature alignment, their quadratic computational complexity makes practical deployment challenging for object detection tasks across diverse domains. Inspired by the global modeling and linear computation complexity of the Mamba architecture, we present the first domain adaptive Mamba-based one-stage object detection model, termed DA-Mamba. Specifically, we combine Mamba's efficient state-space modeling with attention mechanisms to address domain-specific spatial and channel-wise variations. Our design leverages domain adaptive spatial and channel-wise scanning within the Mamba block to extract highly transferable representations for efficient sequential processing, while cross-attention modules generate long-range, mixed-domain spatial features to enable robust soft alignment across domains. Besides, motivated by the observation that hybrid architectures introduce feature noise in domain adaptation tasks, we propose an entropy-based knowledge distillation framework with margin ReLU, which adaptively refines multi-level representations by suppressing irrelevant activations and aligning uncertainty across source and target domains. Finally, to prevent overfitting caused by the mixed-up features generated through cross-attention mechanisms, we propose entropy-driven gating attention with random perturbations that simultaneously refine target features and enhance model generalization. Extensive experiments demonstrate that DA-Mamba consistently outperforms existing methods across a range of widely recognized domain adaptation benchmarks. Our code is available at https://github.com/enesdoruk/DA-Mamba. © 2025 The Authors.Publisher versio