A5101560431- Discourse Analysis and Cultural Communication
- Anomaly Detection Techniques and Applications
- Magnetic and transport properties of perovskites and related materials
- Shape Memory Alloy Transformations
- Traffic Prediction and Management Techniques
- Language, Communication, and Linguistic Studies
- Internet Traffic Analysis and Secure E-voting
- Rare-earth and actinide compounds
- Complex Network Analysis Techniques
- Video Surveillance and Tracking Methods
- Digital Media Forensic Detection
- Network Traffic and Congestion Control
- Advanced Steganography and Watermarking Techniques
- RNA and protein synthesis mechanisms
- Spacecraft Design and Technology
- Magnetic Properties of Alloys
- Network Security and Intrusion Detection
- Advanced Graph Neural Networks
- Cultural, Linguistic, Economic Studies
- Topic Modeling
- Brain Tumor Detection and Classification
- Advanced biosensing and bioanalysis techniques
- DNA and Nucleic Acid Chemistry
- Humor Studies and Applications
- Digital Marketing and Social Media
Hefei University of Technology
2022-2025
Zhengzhou University
2025
Guizhou University
2024
Heilongjiang Bayi Agricultural University
2024
Dalian University
2024
North China University of Science and Technology
2023
Nanjing University of Aeronautics and Astronautics
2022-2023
Northwestern Polytechnical University
2022-2023
Dalian Polytechnic University
2022
Children's Hospital of Capital Institute of Pediatrics
2022
Machine learning has huge potential to revolutionize the field of drug discovery and is attracting increasing attention in recent years. However, lacking domain knowledge (e.g., which tasks work on), standard benchmarks data preprocessing pipelines are main obstacles for machine researchers this domain. To facilitate progress discovery, we develop TorchDrug, a powerful flexible platform built on top PyTorch. TorchDrug variety important including molecular property prediction, pretrained...
Due to network operation and maintenance relying heavily on traffic monitoring, matrix analysis has been one of the most crucial issues for management related tasks. However, it is challenging reliably obtain precise measurement in computer networks because high cost, unavoidable transmission loss. Although some methods proposed recent years allowed estimating from partial flow-level or link-level measurements, they often perform poorly estimation nowadays. Despite strong assumptions like...
Abstract The operational stability issues of perovskite devices under illumination pose significant challenges for practical applications, attributed to substantial alterations in lattice distortion, defect enrichment, and stress distribution the excited state. However, most existing modification strategies predominantly focus on elimination chemical transformations at ground state level, which may not yield desired outcomes To address this challenge, study introduces a dynamic management...
Current protein language models (PLMs) learn representations mainly based on their sequences, thereby well capturing co-evolutionary information, but they are unable to explicitly acquire functions, which is the end goal of representation learning. Fortunately, for many proteins, textual property descriptions available, where various functions also described. Motivated by this fact, we first build ProtDescribe dataset augment sequences with text and other important properties. Based dataset,...
Unsupervised anomaly detection for multivariate time series (MTS) is a challenging task due to the difficulties of precisely learning complex data patterns MTS. The recent progress in sample generation achieved by diffusion models (DMs) motivates us leverage powerful ability DMs make breakthrough unsupervised In this paper, we first attempt design novel diffusion-based model (named TimeADDM) MTS using effective mechanism DMs. To enhance effect on data, propose apply steps representations...
The accuracy and consistency of bone age assessments (BAA) using standard methods can vary with physicians' level experience.To assess the impact information from an artificial intelligence (AI) deep learning convolutional neural network (CNN) model on BAA, specialists different levels experience (junior, mid-level, senior) assessed radiographs 316 children aged 4-18 years that had been randomly divided into two equal sets-group A group B. Bone (BA) was independently by each specialist...
Reasoning on large-scale knowledge graphs has been long dominated by embedding methods. While path-based methods possess the inductive capacity that embeddings lack, their scalability is limited exponential number of paths. Here we present A*Net, a scalable method for graph reasoning. Inspired A* algorithm shortest path problems, our A*Net learns priority function to select important nodes and edges at each iteration, reduce time memory footprint both training inference. The ratio selected...
Researching the cycle level behavior of a processor running an application is crucial to understanding modern computer architecture. To this end, detailed simulators are typically employed. In paper, we present co-simulation framework for out-of-order processors, called QEMU-based sim-outorder, or QSim short. It's combination QEMU and in which does functional simulation, simulation result required by sim-outorder clock-accurate simulation. addition with fast emulation speed QEMU, TB caching...
Nucleic acids, as important substances for biological inheritance, have attracted extensive attention in the biomedical field. More and more cyanine dyes are emerging one of probe tools nucleic acid detection due to their excellent photophysical properties. Here, we discovered that insertion AGRO100 sequence can specifically disrupt twisted intramolecular charge transfer (TICT) mechanism trimethine dye (TCy3), resulting a clear "turn-on" response. Moreover, fluorescence enhancement TCy3...
To perform indicator selection and verification for the on-orbit fault reconstruction of a giant satellite swarm, hybrid multi-objective algorithm is proposed then verified by Monte Carlo analysis. First, according to failure analysis several optimization indicators, such as health state total energy consumption reconstruction, balance fuel consumption, are proposed. Then, fitness function constructed, genetic used optimize objective obtain optimal strategy. Finally, statistically The...
To investigate the changes of main ingredients in black garlic (BG) during fermentation, contents moisture, total acids and reducing sugars were determined. Allicin, 5-Hydroxymethylfurfural (5-HMF), phenols also determined as bioactive substances. DPPH scavenging capacity was to indicate antioxidant activity BG. The hardness color detected well. results showed that moisture content decreased from 66.13% 25.8% fermentation. acids, phenols, increased 0.03 g/g 0.29 g/g, 0.045 μg/g 0.117 μg/g,...
Real-time health assessments are of great importance for the safe and stable operation in-orbit swarms. To solve problems existing real-time microsatellite swarms, such as difficulty selecting a multisource assessment calculation normalization, this paper proposes method applicable to mission-oriented The divides swarm into three levels: single satellite, intersatellite communication link effectiveness, which establish multilevel index system by adopting reliability evaluation based on...
To tackle the issues of low seeding accuracy and seed injury caused by seeders utilized at a small scale in plot sorghum mountainous or hilly regions, this study presents design an oil–electric hybrid air suction seeder. The main working parts seeder are described, performance seed-mixing device is simulated using EDEM software. An drive mode used to provide power for operation seed-metering fan. Additionally, sowing control monitoring system designed single-chip microcomputer controller...
The impact of Zr doping on the characteristics Ni-Mn-In alloy is thoroughly examined in this study. With content no more than 0.5 at.%, introduction widely dispersed second phases results enhanced mechanical properties, allowing a larger volume fraction martensitic transformation during loading. This improvement, combined with decreased interval between temperatures and Curie temperatures, facilitates an increased proportion matrix transitioning from parent phase to martensite while It not...
The impact of Zr doping on the characteristics Ni-Mn-In alloy is thoroughly examined in this study. With content no more than 0.5 at.%, introduction widely dispersed second phases results enhanced mechanical properties, allowing a larger volume fraction martensitic transformation during loading. This improvement, combined with decreased interval between temperatures and Curie temperatures, facilitates an increased proportion matrix transitioning from parent phase to martensite while It not...
Abstract Despite growing attention to the global dissemination of cross-cultural literature, rigorous research on cross-linguistic sentiment analysis remains limited, especially for works like Wolf Totem, which hold both worldwide reach and cultural depth. This study conducted a cross-language thematic online book reviews by Chinese English readers, utilizing advanced linguistic techniques such as BERT, MAXQDA, ProtAnt. The reveals that readers are more critical themes “Ethnicity Culture”,...
Estimating the frequency of items on high-volume, fast data stream has been extensively studied in many areas, such as database and network measurement. Traditional sketch algorithms only allow to give very rough estimates with limited memory cost, whereas some learning-augmented have proposed recently, their offline framework requires actual frequencies that are challenging access general for training, speed is too slow real-time processing, despite still coarse-grained accuracy. To this...