Search

There are now over 20 commercial vector database management systems (VDBMSs), all produced within the past five years. But embedding-based retrieval has been studied for over ten years, and similarity search a staggering half century and more. Driving this shift from algorithms to systems are new data intensive applications, notably large language models, that demand vast stores of unstructured data coupled with reliable, secure, fast, and scalable query processing capability. A variety of new data management techniques now exist for addressing these needs, however there is no comprehensive survey to thoroughly review these techniques and systems. We start by identifying five main obstacles to vector data management, namely the ambiguity of semantic similarity, large size of vectors, high cost of similarity comparison, lack of structural properties that can be used for indexing, and difficulty of efficiently answering "hybrid" queries that jointly search both attributes and vectors. Overcoming these obstacles has led to new approaches to query processing, storage and indexing, and query optimization and execution. For query processing, a variety of similarity scores and query types are now well understood; for storage and indexing, techniques include vector compression, namely quantization, and partitioning techniques based on randomization, learned partitioning, and "navigable" partitioning; for query optimization and execution, we describe new operators for hybrid queries, as well as techniques for plan enumeration, plan selection, distributed query processing, data manipulation queries, and hardware accelerated query execution. These techniques lead to a variety of VDBMSs across a spectrum of design and runtime characteristics, including "native" systems that are specialized for vectors and "extended" systems that incorporate vector capabilities into existing systems. We then discuss benchmarks, and finally outline research challenges and point the direction for future work. [ABSTRACT FROM AUTHOR]

Modern machine learning (ML) systems commonly use stochastic gradient descent (SGD) to train ML models. However, SGD relies on random data order to converge, which usually requires a full data shuffle. For in-DB ML systems and deep learning systems with large datasets stored on block-addressable secondary storage such as HDD and SSD, this full data shuffle leads to low I/O performance—the data shuffling time can be even longer than the training itself, due to massive random data accesses. To balance the convergence rate of SGD (which favors data randomness) and its I/O performance (which favors sequential access), previous work has proposed several data shuffling strategies. In this paper, we first perform an empirical study on existing data shuffling strategies, showing that these strategies suffer from either low performance or low convergence rate. To solve this problem, we propose a simple but novel two-level data shuffling strategy named CorgiPile, which can avoid a full data shuffle while maintaining comparable convergence rate of SGD as if a full shuffle were performed. We further theoretically analyze the convergence behavior of CorgiPile and empirically evaluate its efficacy in both in-DB ML and deep learning systems. For in-DB ML systems, we integrate CorgiPile into PostgreSQL by introducing three new physical operators with optimizations. For deep learning systems, we extend single-process CorgiPile to multi-process CorgiPile for the parallel/distributed environment and integrate it into PyTorch. Our evaluation shows that CorgiPile can achieve comparable convergence rate with the full-shuffle-based SGD for both linear models and deep learning models. For in-DB ML with linear models, CorgiPile is 1.6 × - 12.8 × faster than two state-of-the-art systems, Apache MADlib and Bismarck, on both HDD and SSD. For deep learning models on ImageNet, CorgiPile is 1.5 × faster than PyTorch with full data shuffle. [ABSTRACT FROM AUTHOR]

A ratiometric SERS aptasensor based on catalytic hairpin self-assembly (CHA) mediated cyclic signal amplification strategy was developed for the rapid and reliable determination of Escherichia coli O157:H7. The recognition probe was synthesized by modifying magnetic beads with blocked aptamers, and the SERS probe was constructed by functionalizing gold nanoparticles (Au NPs) with hairpin structured DNA and 4-mercaptobenzonitrile (4-MBN). The recognition probe captured E. coli O157:H7 specifically and released the blocker DNA, which activated the CHA reaction on the SERS probe and turned on the SERS signal of 6-carboxyl-x-rhodamine (ROX). Meanwhile, 4-MBN was used as an internal reference to calibrate the matrix interference. Thus, sensitive and reliable determination and quantification of E. coli O157:H7 was established using the ratio of the SERS signal intensities of ROX to 4-MBN. This aptasensor enabled detection of 2.44 × 102 CFU/mL of E. coli O157:H7 in approximately 3 h without pre-culture and DNA extraction. In addition, good reliability and excellent reproducibility were observed for the determination of E. coli O157:H7 in spiked water and milk samples. This study offered a new solution for the design of rapid, sensitive, and reliable SERS aptasensors. [ABSTRACT FROM AUTHOR]

Based on the criterion of minimum error entropy, this paper proposes a novel conjugate gradient algorithm, called MEE-CG. This algorithm has robust performance under non-Gaussian interference. Theoretical analysis and experimental results demonstrate that the proposed algorithm displays more robust performance than the conventional conjugate gradient methods on the basis of the mean square error and the maximum correntropy criterion. Compared with the stochastic gradient minimum error entropy algorithm and the recursive minimum error entropy algorithm, the proposed algorithm provides a trade-off between computational complexity and convergence speed. [ABSTRACT FROM AUTHOR]

Introduction: Despite the appearance of off-pump coronary artery bypass grafting (CABG), ischemia–reperfusion injury (IRI) in the perioperative period still arouses concerns of clinicians. Remote ischemic conditioning (RIC) is the process of repeated ischemia and reperfusion in the peripheral vessels, which is proven to reduce IRI in vital organs. However, the effect of RIC in patients undergoing off-pump CABG is still unclear. Methods: This IMPROVE trial is a national, multicenter, randomized, controlled, open-label, blinded-endpoint clinical trial designed to assess whether RIC intervention can improve short-term prognosis of patients undergoing off-pump CABG. It plans to enroll 648 patients who will be randomly assigned into a RIC group or control group. Patients in the RIC group will receive four cycles of 5 min of pressurization (about 200 mmHg) and 5 min of rest in the 3 days before and 7 days after the surgery. Planned Outcomes: The primary outcome is the occurrence of major adverse cardiovascular and cerebrovascular events (MACCE) within the 3-month follow-up. MACCE is defined as all-cause death, myocardial infarction, stroke, and coronary revascularization surgery. Clinical Trial Registration: NCT06141525 (ClinicalTrials.gov). [ABSTRACT FROM AUTHOR]

Identifying strong lensing gravitational wave (SLGW) events is of utmost importance in astrophysics as we approach the historic first detection of SLGW amidst the growing number of gravitational wave (GW) events. Currently, one crucial method for identifying SLGW signals involves assessing the overlap of parameters between two GWs. However, the distribution of discrete matter, such as stars and sub-halos, within the strong lensing galaxy can imprint a wave optical (WO) effect on the SLGW waveform. These frequency dependent imprints introduce biases in parameter estimation and impact SLGW identification. In this study, we assess the influence of the stellar microlensing field embedded in a strong lensing galaxy. Our findings demonstrate that the WO effect reduces the detection efficiency of SLGW by 5%–50% for various false alarm probabilities per pair (FAPper pair). Specifically, at an FAPper pair of 10−5, the detection efficiency decreases from ∼ 10% to ∼ 5%. Consequently, the presence of the microlensing field can result in missing half of the strong lensing candidates. Additionally, the microlensing WO effect introduces a noticeable bias in intrinsic parameters, particularly for chirp mass and mass ratio. However, it has tiny influence on extrinsic parameters. Considering all parameters, ∼ 30% of events exhibit a 1σ parameter bias, ∼ 12% exhibit a 2σ parameter bias, and ∼ 5% exhibit a 3σ parameter bias. [ABSTRACT FROM AUTHOR]

The substantial amount of mercury emissions from coal-fired flue gas causes severe environmental contamination. With the Minamata Convention now officially in force, it is critical to strengthen mercury pollution control. Existing activated carbon injection technologies suffer from poor desulfurization performance and risk secondary-release risks. Therefore, considering the potential industrial application of adsorbents, we selected cost-effective and readily available activated coke (AC) as the carrier in this study. Four metal selenides—copper, iron, manganese, and tin—were loaded onto the AC to overcome the application problems of existing technologies. After 120 min of adsorption, the CuSe/AC exhibited the highest efficiency in removing Hg0, surpassing 80% according to the experimental findings. In addition, the optimal adsorption temperature window was 30–120 °C, the maximum adsorption rate was 2.9 × 10−2 mg·g−1·h−1, and the effectiveness of CuSe/AC in capturing Hg0 only dropped by 5.2% in the sulfur-containing atmosphere. The physicochemical characterization results indicated that the AC surface had a uniform loading of CuSe with a nanosheet structure resembling polygon and that the Cu-to-Se atomic ratio was close to 1:1. Finally, two possible Hg0 reaction pathways on CuSe/AC were proposed. Moreover, it was elucidated that the highly selective binding of Hg0 with ligand-unsaturated Se− was the key factor in achieving high adsorption efficiency and sulfur resistance in the selenium-functionalized AC adsorbent. This finding offers substantial theoretical support for the industrial application of this adsorbent. [ABSTRACT FROM AUTHOR]

Coal-fired power plants (CFPPs) are one of the most significant sources of mercury (Hg) emissions certified by the Minamata Convention, which has attracted much attention in recent years. In this study, we used the Web of Science and CiteSpace to analyze the knowledge structure of this field from 2000 to 2022 and then reviewed it systematically. The field of Hg emissions from coal-fired power plants has developed steadily. The research hotspots can be divided into three categories: (1) emission characterization research focused on speciation changes and emission calculations; (2) emission control research focused on control technologies; (3) environmental impact research focused on environmental pollution and health risk. In conclusion, using an oxygen-rich atmosphere for combustion and installing high-efficiency air pollution control devices (APCDs) helped to reduce the formation of Hg0. The average Hg removal rates of APCDs and modified adsorbents after ultra-low emission retrofit were distributed in the range of 82–93% and 41–100%, respectively. The risk level of Hg in combustion by-products was highest in desulfurization sludge (RAC > 10%) followed by fly ash (10% < RAC < 30%) and desulfurization gypsum (1% < RAC < 10%). Additionally, we found that the implementation of pollution and carbon reduction policies in China had reduced Hg emissions from CFPPs by 45% from 2007 to 2015, increased the efficiency of Hg removal from APCDs to a maximum of 96%, and reduced global transport and health risk of atmospheric Hg. The results conjunctively achieved by CiteSpace, and the literature review will enhance understanding of CFPP Hg emission research and provide new perspectives for future research. [ABSTRACT FROM AUTHOR]

Cardinality estimation is one of the most important problems in query optimization. Recently, machine learning-based techniques have been proposed to effectively estimate cardinality, which can be broadly classified into query-driven and data-driven approaches. Query-driven approaches learn a regression model from a query to its cardinality, while data-driven approaches learn a distribution of tuples, select some samples that satisfy a SQL query, and use the data distributions of these selected tuples to estimate the cardinality of the SQL query. As query-driven methods rely on training queries, the estimation quality is not reliable when there are no high-quality training queries, while data-driven methods have no such limitation and have high adaptivity. In this work, we focus on data-driven methods. A good data-driven model should achieve three optimization goals. First, the model needs to capture data dependencies between columns and support large domain sizes (achieving high accuracy). Second, the model should achieve high inference efficiency, because many data samples are needed to estimate the cardinality (achieving low inference latency). Third, the model should not be too large (achieving a small model size). However, existing data-driven methods cannot simultaneously optimize the three goals. To address the limitations, we propose a novel cardinality estimator FACE , which leverages the normalizing flow-based model to learn a continuous joint distribution for relational data. FACE can transform a complex distribution over continuous random variables into a simple distribution (e.g., multivariate normal distribution) and use the probability density to estimate the cardinality for both sequential queries and parallel queries. First, we design a dequantization method to make data more "continuous." Second, we propose encoding and indexing techniques to handle Like predicates for string data. Third, we propose a Monte Carlo method to estimate the cardinality based on the FACE model. Fourth, we propose a grouping technique to process parallel queries. Fifth, we discuss how to support join queries. Experimental results show that our method significantly outperforms existing approaches in terms of estimation accuracy while keeping similar latency and model size. [ABSTRACT FROM AUTHOR]

The proliferation of big data has brought an urgent demand for privacy-preserving data publishing. Traditional solutions to this demand have limitations on effectively balancing the trade-off between privacy and utility of the released data. To address this problem, the database community and machine learning community have recently studied a new problem of tabular data synthesis using generative adversarial networks (GANs) and proposed various algorithms. However, a comprehensive comparison between GAN-based methods and conventional approaches is still lacking, making it unclear why and how GANs can outperform conventional approaches in synthesizing tabular data. Moreover, it is difficult for practitioners to understand which components are necessary when building a GAN model for tabular data synthesis. To bridge this gap, we conduct a comprehensive experimental study that investigates applying GAN to tabular data synthesis. We introduce a unified GAN-based framework and define a space of design solutions for each component in the framework, including neural network architectures and training strategies. We provide optimization techniques to handle difficulties in training GAN in practice. We conduct extensive experiments to explore the design space, comparing with traditional data synthesis approaches. Through extensive experiments, we find that GAN is very promising for tabular data synthesis and provide guidance for selecting appropriate design choices. We also point out limitations of GAN and identify future research directions. We make all code and datasets public for future research. [ABSTRACT FROM AUTHOR]

Double emulsions (DEs) have attracted wide attention because of its excellent encapsulation of hydrophilic and hydrophobic bioactives simultaneously. Nevertheless, the uncontrollable formation and poor stability of DEs limit their further application. Microfluidic technology is a novel strategy to produce DEs with exceptional monodispersity, excellent uniformity, configurable sizes, and adjustable internal droplet numbers. This review mainly summarized the advances of generation and stabilization of DEs based on the microfluidic strategy. We comprehensively focused on the fabrication of DEs under four different types of microfluidic chips and two common curing methods (in situ ultraviolet polymerization and in situ gelation) for DE stabilization, as well as the potential applications in bioactive loading. Also, the problem of efficient generation of microfluidic-produced droplets was raised, and the corresponding solutions were proposed. It could provide theoretical guidance on the development of DE formation and application based on the microfluidic strategy in the food industry. [ABSTRACT FROM AUTHOR]

Large language models (LLMs) have shown superior performance in various areas. And LLMs have the potential to revolutionize data management by serving as the "brain" of next-generation database systems. However, there are several challenges that utilize LLMs to optimize databases. First, it is challenging to provide appropriate prompts (e.g., instructions and demonstration examples) to enable LLMs to understand the database optimization problems. Second, LLMs only capture the logical database characters (e.g., SQL semantics) but are not aware of physical characters (e.g., data distributions), and it requires to fine-tune LLMs to capture both physical and logical information. Third, LLMs are not well trained for databases with strict constraints (e.g., query plan equivalence) and privacy-preserving requirements, and it is challenging to train database-specific LLMs while ensuring database privacy. To overcome these challenges, this vision paper proposes a LLM-based database framework (DB-GPT), including automatic prompt generation, DB-specific model fine-tuning, and DB-specific model design and pre-training. Preliminary experiments show that DB-GPT achieves relatively good performance in database tasks like query rewrite and index tuning. The source code and datasets are available at github.com/TsinghuaDatabaseGroup/DB-GPT. [ABSTRACT FROM AUTHOR]

The increasing number of plant mitochondrial DNA genomes (mtDNA) sequenced reveals the extent of transfer from both chloroplast DNA genomes (cpDNA) and nuclear DNA genomes (nDNA). This study created a library and assembled the chloroplast and mitochondrial genomes of the leafy sweet potato better to understand the extent of mitochondrial and chloroplast gene transfer. The full-length chloroplast genome of the leafy sweet potato (OM808940) is 161,387 bp, with 132 genes annotated, including 87 protein-coding genes, 8 rRNA genes, and 37 tRNA genes. The mitochondrial genome (OM808941) was 269,578 bp in length and contained 69 functional genes, including 39 protein-coding genes, 6 rRNA genes, and 24 tRNA genes. 68 SSR loci were found in the leafy sweet potato organelle genome, including 54 in the chloroplast genome and 14 in the mitochondria genome. In the sweet potato mitochondrial genome, most genes have RNA editing sites, and the conversion ratio from hydrophilic amino acids to hydrophobic amino acids is the highest, reaching 47.12%. Horizontal transfer occurs in the sweet potato organelle genome and nuclear genome. 40 mitochondrial genome segments share high homology with 14 chloroplast genome segments, 33 of which may be derived from chloroplast genome horizontal transfer. 171 mitochondrial genome sequences come from the horizontal transfer of nuclear genome. The phylogenetic analysis of organelle genes revealed that the leafy sweet potato was closely related to the tetraploid wild species Ipomoea tabascana and the wild diploid species Ipomoea trifida. [ABSTRACT FROM AUTHOR]

Here, an enzyme-free lateral flow aptasensor was designed by target-induced strand-displacement effect and followed by the activation of multi-component nucleic acid enzyme (MNAzyme)-mediated cleavage to enable rapid and portable ochratoxin A (OTA) detection. The substrate was prepared as an oligonucleotide strand modified with magnetic beads (MB) and human chorionic gonadotropin (hCG). The interaction of OTA with the aptamer induces the release of blocking DNA, which hybridized with three separated subunits of DNA, forming a sequence-specific MNAzyme catalytic core. This core subsequently initiated an enzyme-free MNAzyme cleavage reaction in the presence of the Mg2+ cofactor, cleaving a special substrate and releasing both the incomplete MNAzyme catalytic core and hCG-DNA probe. The incomplete MNAzyme catalytic core was then recognized by substrates once again, triggering a cascade recycling cleavage and resulting in the generation of a larger number of hCG-DNA probes. After magnetic enrichment, the free hCG-DNA probes flow through the pregnancy test strip (PTS) to the T line, generating a colorimetric readout that unequivocally confirms the presence of the target OTA. This work leverages the efficient enzyme-free cleavage amplification of MNAzyme and the PTS-based portable detection device, presenting a biosensing strategy with significant potential for sensitive and portable OTA detection. This method exhibited remarkable sensitivity and selectivity for OTA detection, boasting a detection limit of 5 nM. The present study successfully demonstrated the practical application of this method on real samples, offering a viable alternative for rapid and portable detection of mycotoxins. [ABSTRACT FROM AUTHOR]

Fluid machinery plays an important role in national pillar industries such as national defense, military, aerospace, heavy industry, energy and power and is also the main industrial energy source. At present, many problems remain in the design of fluid machinery and systems. From the point of view of system optimization design, the utilization rate of fluid mechanics is low, which is mainly due to the mismatch between the system and the network. Based on this, relevant scholars have proposed changing high-pressure fluid machinery hydraulic systems into medium-pressure or low-pressure impeller systems and using hydraulic coupling to regulate the flow of mechanical pumps and fluid machinery to further achieve fluid machinery purposes. However, with the high efficiency, high precision and scalability of fluid machinery, a traditional single core processor has been unable to meet the high precision requirements. Therefore, how to develop software suitable for high-performance computing according to different actual conditions is an important problem. In recent years, with the development of machine learning technology, various algorithms have emerged and been widely used in fluid mechanics. These include the naive Bayesian classifier algorithm, K-means clustering algorithm, K-means clustering machine learning algorithm and support vector machine learning algorithm. Machine learning algorithms based on deep learning have a strong inductive learning ability. They can find the potential flow field information through a large number of experiments and numerical simulations. Through machine learning, a mathematical hydrodynamics gas optimization model is established. The research results show that on this basis, the overall pressure of the hydrodynamic gas optimization model established using a machine learning method was 10.1% higher than that of the control group. This showed that the hydrodynamic gas optimization model established using a machine learning method had better aerodynamic characteristics, providing a reference for future related work. [ABSTRACT FROM AUTHOR]

For China, the development of low-energy buildings is one of the necessary routes for achieving carbon neutrality. Combining photovoltaic (PV) with air source heat pump (ASHP) yields a great potential in providing heating and domestic hot water (DHW) supply in non-central heating areas. However, the diurnal and seasonal inconsistencies between solar availability and building heat load can severely affect the efficacy of solar energy systems. This study creates and numerically simulates a PV-ASHP system with thermal energy storage (TES) in transient system simulation software, TRNSYS. Experimental studies are conducted to validate the simulation model. The system's yearly operational characteristics are simulated to reveal the energy conversion relationship between the system's thermoelectric storage and heating and DHW demand. The results show that the synergy between heating and DHW simultaneously improves the direct utilization of solar energy compared to single heating. The yearly self-consumption and self-satisfaction rates of PV and the COP of the ASHP increase by 131.25%, 10.53% and 9.56%, respectively. Solar energy contributes 55.54% to the system, with a PV capacity of 82 W per square meter of building area. This study provides fresh approaches to developing flexible building-integrated PV-ASHP technologies and balance of the energy exchange among the PV, building load and TES. [ABSTRACT FROM AUTHOR]

Background: This study aimed to summarize the clinical features of Autoimmune Glial Fibrillary Acidic Protein Astrocytosis mimicking tuberculosis meningitis to improve clinicians' understanding of this disease. Methods: We retrospectively analyzed the clinical manifestations, cerebrospinal fluid results, and imaging data of five patients with Autoimmune Glial Fibrillary Acidic Protein Astrocytosis mimicking tuberculous meningitis who were admitted to Xiangya Hospital Central South University between October 2021 and July 2022. Results: Five patients were aged 31–59 years, with a male-to-female ratio of 4:1. Among the cases reviewed, four had a history of prodromal infections manifesting as fever and headache. One patient developed limb weakness and numbness with clinical manifestations of meningitis, meningoencephalitis, encephalomyelitis, or meningomyelitis. Cerebrospinal fluid analysis revealed an increased cell count in five cases, with a lymphocyte majority. All five cases had a CSF protein level > 1.0 g/L, CSF/blood glucose ratio < 0.5, and two patients had CSF glucose < 2.2 mmol/L. Decreased CSF chloride was observed in three cases, while increased ADA was observed in one case. Both serum and cerebrospinal fluid were positive for anti-GFAP antibodies in three cases, while in two cases, only CSF was positive for anti-GFAP antibodies. Additionally, hyponatremia and hypochloremia were observed in three cases. No tumors were detected in any of the five patients during tumor screening, and all five cases had a good prognosis following immunotherapy. Conclusion: Anti-GFAP antibody testing should be routinely performed in patients with suspected tuberculosis meningitis to avoid misdiagnosis. [ABSTRACT FROM AUTHOR]

Despite increasing interest in the study of circulating tumor cells (CTC) subsets, especially epithelial-mesenchymal transition (EMT) and stem cells subsets of CTC that play a key role in tumor recurrence and metastasis, there is no evidence from meta-analyses that shows the correlation between stem-like CTCs and prognosis in cancer patients. Thus, we performed a meta-analysis to assess its prognostic value. Sixteen articles were screened by searching the PubMed, Embase, Cochrane, China National Knowledge Internet (CNKI) and Wanfang databases. The hazard ratio (HR) and 95% confidence interval (95% CI) extracted from each article were summarized. Patients with positive stem-like CTCs in peripheral blood had significantly shorter overall survival (OS, HR: 2.58, 95% CI 1.76–3.79, P < 0.00001), progression-free survival (PFS, HR: 2.21, 95% CI 1.26–3.89, P = 0.006) and disease-free survival (DFS, HR: 2.53, 95% CI: 1.12–5.70, P = 0.03). This study provides the first meta-analysis evidence for the prognostic value of stem-like CTCs, demonstrating that these cells are associated with poor prognosis in cancer patients. Systematic review registration CRD42022322062 [ABSTRACT FROM AUTHOR]

The Sejila tunnel, part of the Sichuan-Tibet railway, is in a complex geostress environment because of its deep burial depths and tectonic movement. Based on the measured stress data combined with the structural history and features, the stress characteristics of the Sejila area are preliminarily identified. Then, a three-dimensional numerical model that can provide real topographic features is established, and a distribution law of stress boundary conditions is proposed according to compilations of much measured stress data. By means of support vector regression (SVR), the stress field of the whole Sejila region is determined and finds a reasonable accordance with the measured stress data. Results show that the vertical stress in deep buried stratum can be approximately regarded as one of the principal stresses, and it is reasonable to apply the lateral stress to the model boundary according to a linear function with burial depth. The in situ stress in the tunnel site exhibits that σH > σV > σh, and the σH direction deflects when it encounters faults or strata interfaces; the larger that the intersecting angle between fault strike and σH is, the smaller the deflection. Compared to the entrance, the rear of the tunnel is subjected to a high maximum principal stress with a high angle; moreover, most sections of the tunnel are estimated to suffer from severe rockbursts, except for a range of 5 km away from the tunnel entrance and 2.5 km away from the exit, according to the Russense criterion. This paper can provide the basis for the prediction and prevention of rockbursts in the Sejila tunnel. [ABSTRACT FROM AUTHOR]

The increase in gravitational wave (GW) events has allowed receiving strong lensing image pairs of GWs. However, the wave effect (diffraction and interference) due to the microlens field contaminates the parameter estimation of the image pair, which may lead to a misjudgment of strong lensing signals. To quantify the influence of the microlens field, researchers need a large sample of statistical research. Nevertheless, due to the oscillation characteristic, the Fresnel-Kirchhoff diffraction integral's computational time hinders this aspect's study. Although many algorithms are available, most cannot be well applied to the case where the microlens field is embedded in galaxy/galaxy clusters. This work proposes a faster and more accurate algorithm for studying the wave optics effect of microlenses embedded in different types of strong lensing images. Additionally, we provide a quantitative estimation criterion for the lens plane boundary for the Fresnel-Kirchhoff diffraction integral. This algorithm can significantly facilitate the study of wave optics, particularly in the case of microlens fields embedded in galaxy/galaxy clusters. [ABSTRACT FROM AUTHOR]

Background: Quantum nanodots especially carbon nanoparticles (CNPs) have been widely studied in biomedicine in imaging, and drug delivery, but anti-cancer mechanisms remain elusive. Methods: Here, we investigated a type of cell death induced by food (beet, soybean) derived CNPs in cancer cells and tested whether CNPs induced DNA damage and resistant to anti-cancer agent PARP inhibitor (PARPi) could be overcome by quantum calculations, TEM, AFM, FT-IR, soft agar assay, and cytotoxicity assay. Results: At high doses, CNPs derived from beet lead to a pop-like apoptosis (Carbopoptosis) in cancer cells. Quantum mechanical calculations confirmed CNPs binding with phosphate groups as well as DNA bases. At low doses, CNPs develop PARPi drug resistance through interactions between CNPs and PARPi. A synergistic drug effect was achieved with the combination of phosphatase inhibitor (PPi), PARPi, and CNPs. This is corroborated by the fact that sulfur modulated CNPs which exhibit super high phosphatase nanozyme activity abrogated the CNPs induced colony formation in anchorage-independent cancer cell growth. Conclusion: Thus, our data suggest the CNPs intrinsic nanozyme activity of phosphatase may crosstalk with drug resistance, which can be reversed upon modulations. [ABSTRACT FROM AUTHOR]

We study the problem of utilizing human intelligence to categorize a large number of objects. In this problem, given a category hierarchy and a set of objects, we can ask humans to check whether an object belongs to a category, and our goal is to find the most cost-effective strategy to locate the appropriate category in the hierarchy for each object, such that the cost (i.e., the number of questions to ask humans) is minimized. There are many important applications of this problem, including image classification and product categorization. We develop an online framework, in which category distribution is gradually learned and thus an effective order of questions are adaptively determined. We prove that even if the true category distribution is known in advance, the problem is computationally intractable. We develop an approximation algorithm, and prove that it achieves an approximation factor of 2. We also show that there is a fully polynomial time approximation scheme for the problem. Furthermore, we propose an online strategy which achieves nearly the same performance guarantee as the offline optimal strategy, even if there is no knowledge about category distribution beforehand. We develop effective techniques to tolerate crowd errors. Experiments on a real crowdsourcing platform demonstrate the effectiveness of our method. [ABSTRACT FROM AUTHOR]

Main conclusion: By constructing an F2 population, a new potential dominant resistance gene to TuMV in Brassica rapa was mapped and identified. Brassica rapa is the most widely grown vegetable crop in China, and turnip mosaic virus (TuMV) is a great threat to its production. Hence, it is a very important work to excavate more and novel resistance genes in B. rapa. In this study, the resistant line B80124 and the susceptible line B80450 were used to construct the F2 populations, and through genetic analysis, the resistance to TuMV was found to be controlled by a dominant gene. Bulked segregant analysis sequence (BSA-seq) was used for the primary mapping, and an intersection (22.25–25.03 Mb) was obtained. After fine mapping using single nucleotide polymorphisms (SNP) markers, the candidate region was narrowed to 330 kb between the SNP markers A06S11 and A06S14, including eight genes relating to disease resistance. Using the transcriptome analysis and sequence identification, BraA06g035130.3C was screened as the final candidate gene, and it contained two deletion mutations, leading to frameshift in the susceptible line B80450. In addition, the phylogenetic analysis, hydrophilia and hydrophobicity analysis, subcellular location prediction analysis, amino acid bias analysis, and 3D modeling structures of BraA06g035130.3C were conducted to predict its functions. This study was conducive to the identification of a new TuMV resistance gene in B. rapa, which is of important scientific significance and application value for the improvement of TuMV resistance traits and molecular design breeding for Brassica crops. [ABSTRACT FROM AUTHOR]

The article titled "Retraction Note: Gray relational clustering model for intelligent guided monitoring horizontal wells" has been retracted by the Editor-in-Chief and the publisher due to concerns regarding compromised editorial handling and peer review process, inappropriate references, and not being in scope of the journal or guest-edited issue. The investigation conducted by the publisher has led to a loss of confidence in the results and conclusions of the article. The authors have not responded to correspondence regarding this retraction. Springer Nature, the publisher, remains neutral in terms of jurisdictional claims and institutional affiliations. [Extracted from the article]

Data exploration—the problem of extracting knowledge from database even if we do not know exactly what we are looking for —is important for data discovery and analysis. However, precisely specifying SQL queries is not always practical, such as "finding and ranking off-road cars based on a combination of Price, Make, Model, Age, Mileage, etc"—not only due to the query complexity (e.g.,the queries may have many if-then-else, and, or and not logic), but also because the user typically does not have the knowledge of all data instances (and their variants). We propose DExPlorer, a system for interactive data exploration. From the user perspective, we propose a simple and user-friendly interface, which allows to: (1) confirm whether a tuple is desired or not, and (2) decide whether a tuple is more preferred than another. Behind the scenes, we jointly use multiple ML models to learn from the above two types of user feedback. Moreover, in order to effectively involve human-in-the-loop, we need to select a set of tuples for each user interaction so as to solicit feedback. Therefore, we devise question selection algorithms, which consider not only the estimated benefit of each tuple, but also the possible partial orders between any two suggested tuples. Experiments on real-world datasets show that DExPlorer outperforms existing approaches in effectiveness. [ABSTRACT FROM AUTHOR]

Main conclusion: Four heterotic QTL and a heterozygous segment for plant weight were identified by Graded Pool-Seq, QTL-seq and traditional genetic linkage analysis in heading Chinese cabbage. Heading Chinese cabbage (Brassica rapa L. spp. pekinensis) is a cross-pollinated leafy vegetable with significant heterosis. The use of heterosis is important for breeding high-yield Chinese cabbage hybrids. However, the formation and mechanism of heterosis have not been studied. We dissected the molecular mechanism of heterosis of yield-related traits in Chinese cabbage. An F1 hybrid with high-parent heterosis of yield-related traits was selected and self-pollinated to generate segregating F2 populations. QTL-seq, Graded Pool-seq (GPS), and traditional genetic linkage analysis were used to identify four heterotic quantitative trait loci (QTL) for plant weight: qPW1.1, qPW5.1, qPW7.1, and qPW8.1. Traditional genetic linkage analysis over two years showed that qPW8.1, located in marker A08_S45 (18,172,719) and A08_S85 (18,196,752), was mapped to a 23.5 kb genomic region. QTL qPW8.1 explained 8.6% and 23.6% of the phenotypic variation in plant weight and the total numbers of head leaves, respectively, and contained a heterozygous segment that might control the heterosis of plant weight. The qPW1.1 made an 11.7% phenotypic contribution to plant weight. The qPW7.1 was sensitive to environmental influence and explained 10.7% of the phenotypic variance. QTL qPW5.1 had a significant signal and was located in a genetic region near the centromere showing high heterozygosity. The “pseudo-overdominance” and “synergistic allelic” effects from parent line “XJD4” appear to play an important role in heterosis for plant weight in Chinese cabbage. These results provide a basis for an improved understanding of the molecular mechanism of yield-related traits and their heterosis. [ABSTRACT FROM AUTHOR]

Lignin is a renewable bioresource that can be used for a variety of value-added applications. However, the effective separation of lignin from lignocellulosic biomass remains an ongoing challenge. In this study, lignin was extracted from waste palm fiber and successfully converted into a dehumidifying material. The following four process parameters of lignin extraction from palm fiber were optimized systematically and comprehensively using the response surface methodology: reaction time, extraction temperature, ethanol concentration and solid/liquid ratio. The results revealed that under the optimum processing conditions (111 min of extraction at 174 °C using 73% ethanol at 1/16 g/mL solid/liquid ratio), the extraction yield of lignin was 56.2%. The recovery of ethanol solvent was as high as 91.8%. Further, the lignin could be directly used without purification to produce lignin-based activated carbon fibers (LACFs) with specific surface area and total pore volume of 1375 m2/g and 0.881 cm3/g, respectively. Compared with the commercial pitch-based activated carbon fiber, the LACF has a higher specific area and superior pore structure parameters. This work provides a feasible route for extracting lignin from natural palm fiber and demonstrates its use in the preparation of activated carbon fiber with a remarkable performance as a solid dehumidification agent. [ABSTRACT FROM AUTHOR]

We systematically investigated the catalytic performance of 3d, 4d, and 5d transition metals anchored onto two-dimensional extended porphyrin (PP) substrates as nitrogen reduction reaction (NRR) electrocatalysts, employing density functional theory (DFT) calculations and four-step high-throughput screening. Four novel metalloporphyrin (MPP, M = Zr, Nb, Hf, and Re) single-atom catalyst candidates have been identified due to their excellent catalytic performance (low onset potential, high stability, and selectivity). Through comprehensive reaction path search, the maximum Gibbs free energy changes for NRR on the ZrPP (enzymatic-consecutive hybrid path), NbPP (consecutive path), HfPP (enzymatic-consecutive hybrid path), and RePP (distal path) catalysts are 0.38, 0.41, 0.53, and 0.53 eV, respectively. Band structures, projected density of states, and charge/spin distributions show that the high catalytic activity is due to significant orbital hybridizations and charge transfer between N2 and MPP catalysts. We hope our work will promote experimental synthesis of these NRR electrocatalysts and provide new opportunities to the electrochemical conversion of N2 to NH3 under ambient conditions. [ABSTRACT FROM AUTHOR]

Computing the shortest paths and shortest path distances between two vertices on road networks is a core operation in many real-world applications, e.g., finding the closest taxi/hotel. However, existing techniques have several limitations. First, traditional Dijkstra-based methods have long latency and cannot meet the high-performance requirement. Second, existing indexing-based methods either involve huge index sizes or have poor performance. To address these limitations, in this paper we propose a learning-based method RNE which can efficiently compute an approximate shortest-path distance such that (1) the performance is super fast, e.g., taking 60–150 nanoseconds; (2) the error ratio of the approximate results is super small, e.g., below 0.7%; (3) scales well to large road networks, e.g., millions of nodes. The key idea is to first embed the road networks into a low dimensional space for capturing the distance relations between vertices, get an embedded vector for each vertex, and then perform a distance metric ( L 1 metric) on the embedded vectors to approximate shortest-path distances. We propose a hierarchical model to represent the embedding, and design an effective method to train the model. We also design a fine-tuning method to judiciously select high-quality training data. In order to identify the shortest path between two vertices (not just the distance), we extend the vertex embedding from RNE and design the RNE+ model, which can output the approximate shortest path with low error and high efficiency. We also propose effective techniques to accelerate the training process of RNE+, including embedding pre-training, negative sampling and model fine-tuning. Extensive experiments on real-world datasets show that RNE and RNE+ significantly outperform the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Chinese cabbage that prefers cold conditions is also affected by low-temperature stress, such as the accumulation of leaf anthocyanins. Research on anthocyanin biosynthesis and regulation mechanisms has made great progress. However, research on anthocyanin accumulation for resistance to biological and non-biological stress is still lacking. To study the relationship between anthocyanin accumulation of Chinese cabbage and resistance under low-temperature conditions, RNA sequencing (RNA-seq) was performed on Chinese cabbage 'Xiao Baojian' grown at a low temperature for four time periods and at a control temperature for five time periods. In Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, 7954 differentially expressed genes (DEGs) were enriched, of which 587 DEGs belonged to "biosynthesis of other secondary metabolites." Gene temporal expression patterns were used to discover enriched genes related to phenylpropanoid biosynthesis; flavonoid biosynthesis and anthocyanin biosynthesis pathways were found in cluster 1. The interaction networks were constructed, and hub genes were selected, showing that flavonoid biosynthesis pathway genes (DFR, ANS, F3H, FLS1, CHS1, CHS3, and TT8) and defense mechanisms-related genes (DFR, SNL6, and TKPR1) interact with each other. Anthocyanin biosynthesis DEGs in Chinese cabbage were evaluated under low-temperature conditions to map the relevant pathways, and expression maps of transcription factors in the flavonoid pathway were created at various periods. Low temperature upregulated the expression of genes related to anthocyanin biosynthesis. Taken together, our results provide further analysis of the relationship between plant anthocyanin synthesis and stress resistance and may also provide further insights for the future development of high-quality color and cold-tolerant Chinese cabbage germplasm resources. [ABSTRACT FROM AUTHOR]

Aim/objective: Aim to evaluate the safety and effectiveness of a pocket compression device in patients undergoing cardiovascular implantable electronic device (CIED) implantation. Material and methods: This randomized controlled trial included 54 patients in the control group (elastic adhesive tape) and 54 patients in the experimental group (pocket compression device). A sandbag was positioned over the pocket after CIED implantation, using elastic adhesive tape in the control group and a novel pocket compression device in the experimental group. The primary outcome was the incidence of pocket hematomas, and the secondary outcomes were adverse skin reactions, the need for sandbag positional adjustments, and patient comfort. Results: The distributions of age, sex, and clinical and procedure characteristics was not significantly different between the two groups (P>0.05). The occurrence of hematomas was lower when using the pocket compression device than in the control group (13.0% vs. 44.4%, P<0.001). The incidence rates of hematomas of grades 1 and the grade 2 hematomas were significantly lower in the experimental group than in the control group (13.0% vs. 31.5%, 0% vs. 11.1% respectively, [all P<0.05]). The incidence rates of adverse skin reactions and sandbag positional adjustments were much lower and the patient comfort score was significantly higher in the experimental group than in the control group (all P<0.001). Conclusion: A pocket compression device can significantly decrease the incidence of pocket hematomas, adverse skin reactions, and positional adjustments when patients are undergoing CIED, which can improve patient comfort and decrease the nursing workload. ClinicalTrials.gov NCT04389398 [ABSTRACT FROM AUTHOR]

Taurine (Tau) is one of the most abundant amino acids in the brain and regulates physiological functions in the central nervous system, including anti-inflammatory effects. There is growing evidence that microglia-mediated neuro-inflammatory responses are an integral part of Parkinson's disease (PD) onset and progression. Among the many factors regulating the inflammatory response, phosphatidylinositol-3 kinase (PI3K) is susceptible to activation by a variety of cytokines and physicochemical factors, and subsequently recruits signaling proteins containing the pleckstrin homology structural domain to further regulate protein kinase B (AKT) expression involved in the regulation of the intracellular immune response and inflammatory response. Therefore, we established a PD mouse model using paraquat (PQ) intraperitoneal injection staining to explore the mechanism of Tau action on PI3K/AKT signaling pathway. Our study showed that PD mice with Tau intervention recovered motor and non-motor functions to some extent, and the number of dopaminergic (DAc) neurons in the substantia nigra and the level of dopamine (DA) secretion in the striatum were also significantly increased compared with the PQ-dyed group, and the protein content of PI3K and PDK-1 and the phosphorylation level of AKT were reduced in parallel with the reduction in the expression of microglia and related inflammatory factors. In conclusion, our results suggest that Tau may regulate microglia-mediated inflammatory responses through inhibition of the PI3K/AKT pathway in the midbrain of PD mice, thereby reducing DAc neurons damage. [ABSTRACT FROM AUTHOR]

Phthalate esters (PAEs) are prone to migrate from the plastic containers into oil during production and storage process. Owing to the trace amount of PAEs and complex sample matrix, it is of great importance to explore a facile and sensitive way to extract and detect PAEs from oil samples. In this work, magnetic covalent organic framework nanospheres (Fe3O4@COF(TbBD)) were fabricated by the aldehyde-amine condensation between benzidine (BD) and 1,3,5-triformylbenzene (Tb) on the surface of Fe3O4 nanoparticles with the help of ultrasonic treatment. The new type nanomaterials were firstly used to the magnetic solid-phase extraction (MSPE) of seven PAEs in edible vegetable oils prior to high performance liquid chromatography coupled with a diode array detector (HPLC–DAD). In order to achieve the best extraction efficiency, several key parameters were optimized. Under optimal conditions, the proposed Fe3O4@COF(TbBD)-MSPE-HPLC–DAD method possessed high sensitivity (0.55–0.95 µg/kg for LODs and 1.80–3.10 µg/kg for LOQs), good linearity (R2 ≥ 0.9983), and satisfactory recoveries (80.2–102.9%). The RSDs for intra-day was less than 6.1%, and inter-day was less than 7.9%. This work not only proposed a rapid and reliable approach for determining PAEs in edible vegetable oils but also broadened the application of magnetic COFs in food determination. [ABSTRACT FROM AUTHOR]

Configuration tuning is vital to optimize the performance of a database management system (DBMS). It becomes more tedious and urgent for cloud databases (CDB) due to diverse database instances and query workloads, which make the job of a database administrator (DBA) very difficult. Existing solutions for automatic DBMS configuration tuning have several limitations. Firstly, they adopt a pipelined learning model but cannot optimize the overall performance in an end-to-end manner. Secondly, they rely on large-scale high-quality training samples which are hard to obtain. Thirdly, existing approaches cannot recommend reasonable configurations for a large number of knobs to tune whose potential values live in such high-dimensional continuous space. Lastly, in cloud environments, existing approaches can hardly cope with the changes of hardware configurations and workloads, and have poor adaptability. To address these challenges, we design an end-to-end automatic CDB tuning system, CDBTune + , using deep reinforcement learning (RL). CDBTune + utilizes the deep deterministic policy gradient method to find the optimal configurations in a high-dimensional continuous space. CDBTune + adopts a trial-and-error strategy to learn knob settings with a limited number of samples to accomplish the initial training, which alleviates the necessity of collecting a massive amount of high-quality samples. CDBTune + adopts the reward-feedback mechanism in RL instead of traditional regression, which enables end-to-end learning and accelerates the convergence speed of our model and improves the efficiency of online tuning. Besides, we propose effective techniques to improve the training and tuning efficiency of CDBTune + for practical usage in a cloud environment. We conducted extensive experiments under 7 different workloads on real cloud databases to evaluate CDBTune + . Experimental results showed that CDBTune + adapts well to a new hardware environment or workload, and significantly outperformed the state-of-the-art tuning tools and DBA experts. [ABSTRACT FROM AUTHOR]

In this paper we investigate the ∞ -capacity and the Faber–Krahn inequality on the Grushin space G α n . On the one hand we show that the ∞ -capacity equals the limit of the p-th root of the p-capacity as p → ∞ and give a simple geometric characterization in terms of the Carnot–Carathéodory distance, on the other hand we establish some basic properties for the ∞ -capacity, ∞ -eigenvalue and the Faber–Krahn inequality on the Grushin space G α n . [ABSTRACT FROM AUTHOR]

Neutral, anionic, and cationic B3Aln0/−/+ (n = 2–6) clusters were systematically explored using density functional theory and coupled cluster CCSD(T) methods to investigate the structural evolution of small mixed aluminum–boron clusters. The lowest energy structures of these clusters were obtained using an unbiased global minimum search, and their structural growth behaviors are discussed. The three boron atoms in B3Aln0/−/+ preferentially form a stable triangle, with additional Al atoms occupying the periphery of the boron triangle. For small clusters of n ≤ 3, the studied clusters show planar two–dimensional configurations. When n ≥ 4, the clusters prefer three–dimensional configurations. Average binding energies, fragmentation energies, second–order differences, HOMO–LUMO gaps, ionization potentials and electron affinities are discussed in detail. For cationic B3Aln+ (n = 2–6) clusters, the even n systems are more stable than the odd n systems, while the stabilities of neutral B3Aln and anionic B3Aln− clusters do not change significantly with growing n. The infrared spectrum and photoelectron spectroscopy of these clusters are simulated, which will be useful for future experimental research. We also compare the chemical bonding of neutral B3Aln (n = 2–6) clusters with their ionic clusters by AdNDP analysis. [ABSTRACT FROM AUTHOR]

Classification of brain tumors based on the brain magnetic resonance imaging (MRI) results of patients has become an important problem in medical image processing. A computer program that can efficiently analyze brain MRI images of patients in real time and generate accurate classification results of the tumors in these images can significantly reduce the amount of time needed for diagnosis, which may increase the chances for patients to survive. This paper proposes a new statistical method that can accurately classify three types of brain tumors based on MRI images, the three types of tumors considered include pituitary tumor, glioma, and meningioma. The features for a pixel in an MRI image are obtained by applying a set of convolutional operators to the neighborhood area of the pixel. For training, a hidden Markov model (HMM) is constructed and trained from a training dataset by computing a statistical profile for the feature vectors for pixels in the tumor regions of each type of brain tumors. In addition, a statistical profile is also obtained for pixels that are in the background of a tumor. For classification, the trained HMM is used to assign labels to pixels in an MRI image with a dynamic programming approach and the classification result of the image is obtained from the labels assigned to the tumor region. Both the training and classification processes can be efficiently performed in linear time and does not require the availability of a large amount of computational resources. Experimental results on a large dataset of MRI images show that the proposed method can provide classification results with high accuracy for all three types of brain tumors. A comparison with state-of-the-art methods for brain tumor classification suggests that the proposed method can achieve improved classification accuracy. In addition, real-time analysis also reveals that the proposed approach can probably be used for real-time classification of brain tumors. [ABSTRACT FROM AUTHOR]

Entity categorization, the process of categorizing entities into groups, is an important problem with many applications. However, in practice, many entities are mis-categorized, such as Google Scholar and Amazon products. In this paper, we study the problem of discovering mis-categorized entities from a given group of categorized entities. This problem is inherently hard: All entities within the same group have been "well" categorized by the state-of-the-art solutions. Apparently, it is nontrivial to differentiate them. We propose a novel rule-based framework to solve this problem. It first uses positive rules to compute disjoint partitions of entities, where the partition with the largest size is taken as the correctly categorized partition, namely the pivot partition. It then uses negative rules to identify mis-categorized entities in other partitions that are dissimilar to the entities in the pivot partition. We describe optimizations on applying these rules and discuss how to generate positive/negative rules. In addition, we propose novel strategies to resolve inconsistent rules. Extensive experimental results on real-world datasets show the effectiveness of our solution. [ABSTRACT FROM AUTHOR]

Main conclusion: Gene co-expression network analysis of the heat-responsive core transcriptome in two contrasting Brassica rapa accessions reveals the main metabolic pathways, key modules and hub genes, are involved in long-term heat stress. Brassica rapa is a widely cultivated and economically important vegetable in Asia. High temperature is a common stress that severely impacts leaf head formation in B. rapa, resulting in reduced quality and production. The purpose of this study was thus to identify candidate heat tolerance genes by comparative transcriptome analysis of two contrasting B. rapa accessions in response to long-term heat stress. Two B. rapa accessions, ‘268’ and ‘334’, which showed significant differences in heat tolerance, were used for RNA sequencing analysis. We identified a total of 11,055 and 8921 differentially expressed genes (DEGs) in ‘268’ and ‘334’, respectively. Functional enrichment analyses of all of the identified DEGs, together with the genes identified from weighted gene co‐expression network analyses (WGCNA), revealed that the autophagy pathway, glutathione metabolism, and ribosome biogenesis in eukaryotes were significantly up-regulated, whereas photosynthesis was down-regulated, in the heat resistance of B. rapa ‘268’. Furthermore, when B. rapa ‘334’ was subjected to long-term high-temperature stress, heat stress caused significant changes in the expression of certain functional genes linked to protein processing in the endoplasmic reticulum and plant hormone signal transduction pathways. Autophagy-related genes might have been induced by persistent heat stress and remained high during recovery. Several hub genes like HSP17.6, HSP17.6B, HSP70-8, CLPB1, PAP1, PYR1, ADC2, and GSTF11 were discussed in this study, which may be potential candidates for further analyses of the response to long-term heat stress. These results should help elucidate the molecular mechanisms of heat stress adaptation in B. rapa. [ABSTRACT FROM AUTHOR]

Internal cracks are one of the main defects limiting the quality of continuous casting blooms. With bearing steel GCr15 chosen as the specific research steel, the temperature of internal crack initiation was determined by the solid-phase fraction (fs) of the fracture, and by a simple and efficient high-temperature tensile method, the critical strain of internal crack initiation was determined. The thermal and mechanical behavior of continuous casting bloom was simulated and analyzed by a three-dimensional (3D) thermomechanical finite element model, and the model was verified by the online measured relationship between reduction force and amount of withdrawal and straightening units. Combining the critical criterion and strain distribution, an internal crack risk prediction model for continuous casting blooms was presented, and the predicted crack risk contour plot was verified by the morphology of actual bloom internal cracks. The maximum allowable single-roll reduction amount with different solidification phase fractions was analyzed for 320 mm × 425 mm GCr15 blooms, which concluded that the reasonable reduction strategy of continuous casting blooms could meet the requirement of relief center segregation and avoid internal crack risk simultaneously. [ABSTRACT FROM AUTHOR]

Intelligent transportation (e.g., intelligent traffic light) makes our travel more convenient and efficient. With the development of mobile Internet and position technologies, it is reasonable to collect spatio-temporal data and then leverage these data to achieve the goal of intelligent transportation, and here, traffic prediction plays an important role. In this paper, we provide a comprehensive survey on traffic prediction, which is from the spatio-temporal data layer to the intelligent transportation application layer. At first, we split the whole research scope into four parts from bottom to up, where the four parts are, respectively, spatio-temporal data, preprocessing, traffic prediction and traffic application. Later, we review existing work on the four parts. First, we summarize traffic data into five types according to their difference on spatial and temporal dimensions. Second, we focus on four significant data preprocessing techniques: map-matching, data cleaning, data storage and data compression. Third, we focus on three kinds of traffic prediction problems (i.e., classification, generation and estimation/forecasting). In particular, we summarize the challenges and discuss how existing methods address these challenges. Fourth, we list five typical traffic applications. Lastly, we provide emerging research challenges and opportunities. We believe that the survey can help the partitioners to understand existing traffic prediction problems and methods, which can further encourage them to solve their intelligent transportation applications. [ABSTRACT FROM AUTHOR]

The structures and properties of binary boron-aluminum B5Aln0/-/+ (n = 1–4) clusters have been systematically explored using the density functional theory method at the B3LYP/6–311+G(d) level and the coupled cluster method at the CCSD(T)/6–311+G(2df)//B3LYP/6–311+G(d) level. Lowest-energy structures, stabilities, growth behaviors and chemical bonding of these clusters were analyzed. Our results show that when the number of doped Al atoms is one or two, the Al atoms are located at the periphery, and the host B5 cluster preferentially forms a W-shape core, which is only slightly affected by the Al atoms. When there are three or four Al atoms, the B5Aln0/-/+ (n = 3, 4) clusters have their lowest energy structures preferentially in capped bipyramid configurations. Neutral B5Aln (n = 1, 3) clusters are somewhat more stable than their neighboring n clusters, while anionic and cationic B5Aln-/+ (n = 2, 4) clusters tend to be somewhat more stable. We also simulated the infrared (IR) spectrum and photoelectron spectroscopy (PES) of these clusters for future experimental comparison. Adaptive natural density partitioning (AdNDP) analysis shows that a variety of delocalized multicenter bonds appear in these clusters, which may enhance the stability of the clusters. [ABSTRACT FROM AUTHOR]

Image enhancement is a problem of fundamental importance in the area of low level image processing. The goal of image enhancement is to significantly improve the visual effects of images or to obtain the fine details that are invisible in degraded images. In this paper, a new accurate image enhancement algorithm is developed to efficiently perform image enhancement with a dynamic programming approach. Specifically, an objective function is developed for the mappings between an original image and its enhanced versions to evaluate the effectiveness of enhancement. The objective function is then optimized by a dynamic programming algorithm to achieve the optimal enhancement effect. It is also shown that the computation efficiency of this dynamic programming algorithm can be significantly improved when certain conditions are satisfied. Testing results show that this new algorithm can efficiently generate images with significantly improved effectiveness of enhancement and is thus potentially useful for real-time applications. An implementation of the algorithm in MATLAB is freely available at the link: https://github.com/yinglei2020/YingleiSong. [ABSTRACT FROM AUTHOR]

Data labeling, which assigns data with multiple classes, is indispensable for many applications, such as machine learning and data integration. However, existing labeling solutions either incur expensive cost for large datasets or produce noisy results. This paper introduces a cost-effective labeling approach and focuses on the labeling rule generation problem that aims to generate high-quality rules to largely reduce the labeling cost while preserving quality. To address the problem, we first generate candidate rules and then devise a game-based crowdsourcing approach CrowdGame to select high-quality rules by considering coverage and accuracy. CrowdGame employs two groups of crowd workers: One group answers rule validation tasks (whether a rule is valid) to play a role of rule generator, while the other group answers tuple checking tasks (whether the label of a data tuple is correct) to play a role of rule refuter. We let the two groups play a two-player game: Rule generator identifies high-quality rules with large coverage, while rule refuter tries to refute its opponent rule generator by checking some tuples that provide enough evidence to reject rules with low accuracy. This paper studies the challenges in CrowdGame. The first is to balance the trade-off between coverage and accuracy. We define the loss of a rule by considering the two factors. The second is rule accuracy estimation. We utilize Bayesian estimation to combine both rule validation and tuple checking tasks. The third is to select crowdsourcing tasks to fulfill the game-based framework for minimizing the loss. We introduce a minimax strategy and develop efficient task selection algorithms. We also develop a hybrid crowd-machine method for effective label assignment under budget-constrained crowdsourcing settings. We conduct experiments on entity matching and relation extraction, and the results show that our method outperforms state-of-the-art solutions. [ABSTRACT FROM AUTHOR]