17315 research outputs found
Sort by
PHACE: phylogeny-aware detection of molecular coevolution
No full textThe coevolution trends of amino acids within or between genes offer key insights into protein structure and function. Existing tools for uncovering coevolutionary signals primarily rely on multiple sequence alignments, often overlooking phylogenetic relatedness and shared evolutionary history. Here, we introduce PHACE, a phylogeny-aware coevolution algorithm that maps amino acid substitutions onto a phylogenetic tree to detect molecular coevolution. PHACE categorizes amino acids at each position into "tolerable"and "intolerable"groups, based on their independent recurrence across the tree, reflecting a position's tolerance to specific substitutions. Gaps are treated as a third character type, with only phylogenetically independent gap changes considered. The method computes substitution scores per branch by traversing the tree and quantifying probability differences across adjacent nodes for each group. To avoid artifacts from alignment errors, we apply a multiple sequence alignment-masking procedure. Compared to phylogeny-based methods (CAPS, CoMap) and state-of-the-art multiple sequence alignment-based approaches (DCA, GaussDCA, PSICOV, mutual information), PHACE shows significantly superior accuracy in identifying coevolving residue pairs, as measured by statistical metrics including Matthews correlation coefficient, area under the ROC curve, and F1 score. This performance stems from PHACE's explicit modeling of phylogenetic dependencies, often ignored in coevolution analyses
Porosity effect of bio-coated surfaces on flow boiling heat transfer of HFE-7000
Full text linkSurface modification has emerged as a prominent approach for providing ultra-high heat flux cooling and enhancing boiling heat transfer of particularly dielectric fluids, where inherent limitations such as low thermal conductivity and latent heat deteriorate the performance. Despite the large number of methodologies in surface modification for enhanced boiling heat transfer, a considerable proportion of surface modification techniques necessitate access to cleanroom facilities or involve protracted procedures, including the utilization of environmentally hazardous materials. This study presents flow boiling heat transfer results of HFE-7000 in a rectangular high aspect ratio minichannel with a coated surface using proposed environmentally friendly and economical microbial bio-coating (Saccharolobus solfataricus P2 bio-coatings). The boiling heat transfer characteristics of bio-coated surfaces, which were optimized in terms of coating structure and durability against the fluid flow by the dip mixed coating method, were explored. Flow boiling experiments were performed on coated surfaces at heat fluxes ranging from 5.4 to 50.9 W/cm2, mass flux of 500 kg/m2s, inlet subcooling of 10 ℃, and atmospheric pressure. Bubble dynamics and flow boiling patterns were obtained using a high-speed camera. The bio-coated surface offered significant heat transfer enhancement compared to the bare silicon sample by offering more active nucleation sites and showing resistance to vapor film/dry spot formation on the surface by providing a porous structure. This porous architecture not only increases the density of nucleation sites but also enables capillary-driven rewetting, which sustains thin liquid films and delays dry-out at high heat fluxes. The coated surface achieved the highest heat transfer coefficient, with a maximum enhancement of 50 %, and enhanced the critical heat flux (by a factor of about 1.5 relative to the plain surface) by stabilizing two-phase flow and promoting rewetting
Text-based image retrieval system using semantic visual content for re-ranking
No full textWe address the problem of retrieving relevant images from the web in response to a text query. Text-based image retrieval is a challenging multimodal problem that requires understanding of the user intent from a few keywords and a semantic understanding of the images to predict relevance. In this paper, we present the hybrid search approach of the Turkcell-Yaani Search Engine, which was the leading search engine for queries in the Turkish language in 2017–2022. With over 1 million monthly users and processing millions of requests per month, its hybrid image search approach demonstrates the practical relevance and effectiveness of integrating text and visual information for enhanced image retrieval. The initial stage of the system focuses on retrieving images by comparing the query text with textual information associated with previously crawled and indexed images. The second stage aims to address the limitations of relying solely on text-based information by incorporating the post-processing steps based on semantic analysis of retrieved images. This stage, which is the main contribution of this paper, involves clustering, filtering and reordering the retrieved images using their embedded representations that are obtained from pretrained deep neural networks. The experimental results show that incorporating visual content improves the performance of the text-based retrieval engine significantly
Improved rank-one-based relaxations and bound tightening techniques for the pooling problem
Full text linkThe pooling problem is a classical NP-hard problem in the chemical process and petroleum industries. This problem is modeled as a nonlinear, nonconvex network flow problem in which raw materials with different specifications are blended in some intermediate tanks and mixed again to obtain the final products with the desired specifications. The analysis of the pooling problem is quite an active research area, and different exact formulations, relaxations, and restrictions are proposed. In this paper, we focus on a recently proposed rank-one-based formulation of the pooling problem. In particular, we study a recurring substructure in this formulation defined by the set of nonnegative, rank-one matrices with bounded row sums, column sums, and the overall sum. We show that the convex hull of this set is second-order cone representable. In addition, we propose an improved compact-size polyhedral outer-approximation and families of valid inequalities for this set. To further strengthen these convexification approaches, we develop two bound tightening techniques that refine the capacities of source, pool, and terminal nodes, as well as flow bounds on arcs. One is a simple, rule-based approach using network structure (e.g., supply, demand, and neighboring capacities), while the other solves auxiliary optimization problems to compute tighter bounds. Our computational experiments show that the newly proposed polyhedral outer-approximation can improve upon the traditional linear programming relaxations of the pooling problem in terms of the dual bound. Furthermore, bound tightening techniques reduce the computational time spent on both the exact method, linear programming, and mixed-integer linear programming relaxations
ThresholdFP: enhanced durability in browser fingerprinting
Full text linkBrowser fingerprinting is a powerful tool for user identification in financial and other security-critical applications that require strong authentication. However, due to the instability of browser attributes, fingerprints often change rapidly, reducing their lifespan and negatively impacting user convenience. We propose ThresholdFP, a novel linking algorithm designed to extend the durability of browser fingerprints without compromising precision. Instead of replacing the fingerprint after even a small change, ThresholdFP computes a difference score between fingerprints and tolerates variations as long as they stay within a predefined threshold. To ensure realistic performance evaluation, we collected and utilized two real-world datasets. Comparative results show that the fingerprints generated by ThresholdFP are significantly more persistent than those produced by existing methods, while maintaining near-perfect precision. Notably, our approach achieves an improvement in average tracking duration ranging from 24.33% to 106.30% compared to rival schemes in the literature
Generalizing the Bierbrauer-Friedman bound for orthogonal arrays
No full textWe characterize mixed-level orthogonal arrays in terms of algebraic designs in a special multigraph. We prove a mixed-level analog of the Bierbrauer–Friedman (BF) bound for pure-level orthogonal arrays and show that arrays attaining it are radius-1 completely regular codes (equivalently, intriguing sets, equitable 2-partitions, perfect 2-colorings) in the corresponding multigraph. For the case when the numbers of levels are powers of the same prime number, we characterize, in terms of multispreads, additive mixed-level orthogonal arrays attaining the BF bound. For pure-level orthogonal arrays, we consider versions of the BF bound obtained by replacing the Hamming graph by its polynomial generalization and show that in some cases this gives a new bound
İktisadi düşünce tarihi notları (1) Wladimir Lenin: Bolşevik Devrimi'nin lideri ve Sovyetler Birliği'nin kurucusu
No full textTool wear prediction in milling process using physics-informed machine learning and thermo-mechanical force model with monitoring applications
No full textAccurate wear estimation of milling tools is critical for enhancing the productivity and reliability of machining processes, ensuring consistent product quality while minimizing unexpected tool failure, downtime and machining costs. Traditional approaches, often based on pure experimental and data-driven machine learning (ML) methods, demand extensive, costly wear testing to gather the necessary datasets, which limits their utility in practical industrial monitoring. To address this gap, this work presents a novel physics-informed machine learning (PIML) approach of wear estimation by integrating analytical models with ML techniques. The PIML model utilizes a wear-inclusive thermo-mechanical model to estimating cutting forces considering flank wear and edge forces, with special focus on its adaptation to milling operations and addressing the complexities of milling dynamics. The methodology is demonstrated on Steel 1050, a widely used medium-carbon steel alloy in industrial machining applications. As shown by the results, this hybrid model shows high predictive accuracy, achieving R² values exceeding 98 % for force prediction and 95 % for tool wear estimation, with corresponding RMSE values below 14 N and 8 µm, respectively. Notably, the use of the PIML framework improved tool wear prediction accuracy by over 16 % compared to using ML alone. Another important finding is the significant role of edge forces under severe wear conditions, with their contribution to average cutting forces increasing from 40 % to 57 % at low feed rates, and from 27 % to 45 % at higher feed rates. Using this enhanced model, a simulation-based dataset was generated to train an inverse ML model for estimating tool wear considering milling forces and cutting parameters. The inverse ML model exhibited robust predictive performance, offering a practical and accurate solution for tool wear estimation. This study emphasizes the promising potential of integrating thermo-mechanical model with ML algorithms in machining applications, establishing a foundation of tool wear condition monitoring through milling force data. The presented approach can contribute to enhanced process control, optimized tool usage, and reduced operational costs. Furthermore, it supports the transition to Industry 4.0 by enabling automation and unsupervised manufacturing, where real-time tool wear monitoring and adaptive control can be achieved with minimal human intervention, driving more intelligent and efficient manufacturing systems
A comprehensive experimental study on the development of high-performance sandwich panels using polyether sulfone (PESU) thermoplastic core and skin for aircraft interior applications
No full textThermoplastics are increasingly used in aircraft sandwich composites due to their recyclability; however, their effective application relies on achieving strong core-to-skin bonding. This work focuses on developing advanced sandwich panels using Polyethersulfone (PESU) thermoplastic to accomplish these goals. The approach involves combining a PESU core with skin layers composed of glass fiber reinforced PESU film (rGF/PESU skin). The materials are processed using lower side hot pressing at different temperatures (265 °C and 270 °C) and durations (45 and 60 s) to optimize their properties. A comprehensive set of chemical and morphological analyses is performed to assess the characteristics of the PESU core, rGF/PESU skin, and the resulting hot-pressed sandwich panels. XRD analyses show that the PESU core exhibits semi-crystalline behavior, which decreases with the addition of amorphous glass fibers and is further reduced under hot press processing due to foam structure disruption. Results demonstrate a significant improvement in thermal performance, with the thermal conductivity of the hot-pressed sandwich panel at 270 °C for 60 s increasing by approximately 145 % compared to the unmodified PESU core, indicating enhanced heat transfer capabilities. The highest flexural strength of 20.1 MPa is attained for three-point bending tests. Advanced imaging techniques, such as computed tomography (CT) scans and Scanning Electron Microscopy (SEM), and Optical Microscopy (OP), reveal three distinct phases within the structure: the PESU core, the rGF/PESU skin, and an interlayer phase that forms between them during hot pressing. Based on the OP analysis, the sandwich panel processed at 270 °C for 60 s exhibits the maximum interlayer transition phase thickness, reaching approximately 500 μm
Thermodynamics of Einstein-Geometric Proca AdS compact objects
Full text linkIn this study, we explore metric-Palatini gravity extended by the antisymmetric component of the affine curvature. This gravitational theory results in general relativity plus a geometric Proca field. Building on our previous work, where we constructed its static spherically symmetric solutions in the Anti-de Sitter (AdS) background (Eur. Phys. J. C 83(4):318, 2023), we conduct a comprehensive analysis of the system’s thermodynamics. We examine the thermodynamic properties of the Einstein-Geometric Proca AdS compact objects, focusing on the Hawking temperature, enthalpy, heat capacity, entropy, and Gibbs free energy. Particular attention is given to the dependence of the Hawking temperature, enthalpy, and heat capacity on the uniform potential q1 and the electromagnetic-type charge q2. Through numerical analysis, we compute the entropy and Gibbs free energy and investigate how these quantities vary with the model parameters