A5055857919- Magnetic confinement fusion research
- Advanced Graph Neural Networks
- Complex Network Analysis Techniques
- Graph Theory and Algorithms
- Fusion materials and technologies
- Data Management and Algorithms
- Particle accelerators and beam dynamics
- Algorithms and Data Compression
- Ionosphere and magnetosphere dynamics
- Superconducting Materials and Applications
- Traffic Prediction and Management Techniques
- Complexity and Algorithms in Graphs
- Web Data Mining and Analysis
- Data Mining Algorithms and Applications
- Caching and Content Delivery
- Laser-Plasma Interactions and Diagnostics
- Advanced Malware Detection Techniques
- Human Mobility and Location-Based Analysis
- Topic Modeling
- Natural Language Processing Techniques
- Plasma Diagnostics and Applications
- Geophysics and Sensor Technology
- Advanced Optical Sensing Technologies
- Automated Road and Building Extraction
- IPv6, Mobility, Handover, Networks, Security
Chinese University of Hong Kong, Shenzhen
2023-2025
Shandong Agricultural University
2024
Changchun University of Science and Technology
2015-2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2024
Peking Union Medical College Hospital
2024
Harvard University
2024
Massachusetts General Hospital
2024
Brigham and Women's Hospital
2024
Yanshan University
2022
University of Hong Kong
2019-2022
Text-to-SQL parsing, which aims at converting natural language instructions into executable SQLs, has gained increasing attention in recent years. In particular, Codex and ChatGPT have shown impressive results this task. However, most of the prevalent benchmarks, i.e., Spider, WikiSQL, focus on database schema with few rows contents leaving gap between academic study real-world applications. To mitigate gap, we present Bird, a big benchmark for large-scale grounded text-to-SQL tasks,...
The task of text-to-SQL parsing, which aims at converting natural language questions into executable SQL queries, has garnered increasing attention in recent years. One the major challenges parsing is domain generalization, i.e., how to generalize well unseen databases. Recently, pre-trained text-to-text transformer model, namely T5, though not specialized for achieved state-of-the-art performance on standard benchmarks targeting generalization. In this work, we explore ways further augment...
Given a directed graph G, the densest subgraph (DDS) problem refers to finding of from whose density is highest among all subgraphs G. The DDS fundamental wide range applications, such as fraud detection, community mining, and compression. However, existing solutions suffer efficiency scalability problems: on three-thousand-edge graph, it takes three days for one best exact algorithms complete. In this paper, we develop an efficient scalable solution. We introduce notion [x, y]-core, which...
In graph applications (e.g., biological and social networks), various analytics tasks clustering community search) are carried out to extract insight from large complex graphs. Central these is the counting of number motifs , which graphs with a few nodes. Recently, researchers have developed several fast motif algorithms. Most solutions assume that deterministic, i.e., edges certain exist. However, due measurement statistical prediction errors, this assumption may not hold, hence analysis...
In this paper, we study anomalous trajectory detection, which aims to extract abnormal movements of vehicles on the roads. This important problem, facilitates understanding traffic behavior and detection taxi fraud, is challenging due varying conditions at different times locations. To tackle propose <u>deep</u> -probabilistic-based <u>t</u>ime-d<u>e</u>pendent <u>a</u>nomaly algorithm ( DeepTEA ). method, employs deep-learning methods obtain...
Road network applications, such as navigation, incident detection, and Point-of-Interest (POI) recommendation, make extensive use of edge weights (e.g., traveling times). Some these can be missing, especially in a road where traffic data may not available for every road. In this paper, we study the stochastic weight completion (SWC) problem, which computes distributions missing edges. This is difficult, due to intricate temporal spatial correlations among neighboring Besides, sparse , i.e.,...
This is the reproducibility report for paper ''Temporal JSON Keyword Search'' by Dyreson et al. The code authors provided available on GitHub https://github.com/cdyreson/temporalHierarchicalKeywordSearch. We tested code's correctness, data availability, and of results. In particular, we aim to replicate two key findings evaluation: cost temporal functionality additional overhead sequenced search compared consequences version. Generally, results are consistent with original paper. Overall,...
Neutral-beam injection of up to 2.5 MW into plasmas in the ISX-B tokamak (R0 = 0.93 m, a 0.27 BT 0.9–1.5 T, Ip 70–210 kA, e 2.5–10×1013 cm−3) has created with volume-averaged beta ∼ 2.5%, peak values 9%, and root-mean-square 3.5%. Energy confinement time is observed decrease by about factor two as beam power goes from 0 MW; caused predominantly electron falling below predictions 'Alcator scaling' 3–4 at high power. An empirical relationship form fits our measurements over wide range plasma...
Community search (CS) enables personalized community discovery and has found a wide spectrum of emerging applications such as setting up social events friend recommendation. While CS been extensively studied for conventional homogeneous networks, the problem heterogeneous information networks (HINs) received attention only recently. However, existing studies suffer from several limitations, e.g., they either require users to specify meta-path or relational constraints, which pose great...
Given a directed graph G, the densest subgraph (DDS) problem refers to finding from whose density is highest among all subgraphs of G. The DDS fundamental wide range applications, such as fake follower detection and community mining. Theoretically, closely connects other essential problems, network flow bipartite matching. However, existing solutions suffer efficiency scalability issues. In this paper, we develop convex-programming-based solution by transforming into set linear programs....
The densest subgraph problem (DSP) is of great significance due to its wide applications in different domains. Meanwhile, diverse requirements various lead density variants for DSP. Unfortunately, existing DSP algorithms cannot be easily extended handle those efficiently and accurately. To fill this gap, we first unify metrics into a generalized definition. We further propose new model, c-core, locate the general show advantage accelerating searching process. Extensive experiments that our...
An experiment to test beryllium as a limiter material has been performed in the ISX-B tokamak. The effect of plasma on and were studied detail. Heat particle fluxes measured, damage by melting was documented function power flux. Strong evaporation caused gettering vacuum vessel. Postmortem analysis document amount retained hydrogen erosion impurity deposition limiter. performance terms parameter space, content, confinement for ungettered gettered cases. Operational experience with fusion is...
Path-based solutions have been shown to be useful for various graph analysis tasks, such as link prediction and clustering. However, they are no longer adequate handling complex gigantic graphs. Recently, motif-based has attracted a lot of attention. A motif, or small with few nodes, is often considered fundamental unit graph. Motif-based captures high-order structure between performs better than traditional "edge-based" solutions. In this paper, we study motif-path , which conceptually...
Given a directed graph G , the densest subgraph (DDS) problem refers to finding of from whose density is highest among all subgraphs . The DDS fundamental wide range applications, such as fraud detection, community mining, and compression. However, existing solutions suffer efficiency scalability problems: on 3,000-edge graph, it takes three days for one best exact algorithms complete. In this article, we develop an efficient scalable solution. We introduce notion [ x y ]-core, which dense...
In this paper, for the first time, we introduce concepts of window-CCs and window-SCCs on undirected directed temporal graphs, respectively. We then study queries window-CC window-SCC by developing several efficient index-based query solutions. The space costs best indices are linear to sizes graphs. extensive experimental evaluation 12 real-world datasets demonstrates high efficiency effectiveness proposed future, will develop distributed index construction algorithms, which would be useful...
Densest subgraph discovery (DSD) is a fundamental topic in graph mining. It has been extensively studied the literature and found many real applications wide range of fields, such as biology, finance, social networks. As typical problem DSD, k-clique densest (CDS) aims to detect from graph, that ratio number k-cliques over its vertices maximized. This received plenty attention literature, widely used identifying larger ''near-cliques''. Existing CDS solutions, either k-core or convex...
Experiments to investigate the scaling of volume-averaged beta <β> and a global energy confinement time for neutral-beam-heated (Pb ≤ 2.5 MW) discharges in ISX-B tokamak are described. The results condensed into set empirical formulas which can be used as guide other theoretical experimental studies high-beta, plasmas. dependence on toroidal field BT, plasma current Ip, line-averaged electron density e was determined by varying each these while keeping external variables fixed. Magnetic...
The effect of periodic toroidal field (TF) ripple on ion confinement has been studied in the ISX-B tokamak by comparing neutral-beam-heated plasma performance with 9 and 18 TF coils. Three physics issues were treated these experiments: (1) enhanced thermal conductivity, (2) loss energetic ions, (3) damping beam-induced rotation, which may affect losses. Under a wide variety conditions, reduced central-ion temperature factor approximately two (600 eV → 300 eV). Ion was found to be nearly...