Mengchen Zhang

ORCID: 0000-0003-3670-8767
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Membrane Separation Technologies
  • Theoretical and Computational Physics
  • Topic Modeling
  • Mobile Crowdsensing and Crowdsourcing
  • Chemical Synthesis and Characterization
  • Statistical Mechanics and Entropy
  • Data Quality and Management
  • Membrane-based Ion Separation Techniques
  • Extraction and Separation Processes
  • Semantic Web and Ontologies
  • Membrane Separation and Gas Transport
  • Stochastic processes and financial applications
  • Stochastic processes and statistical mechanics
  • Random Matrices and Applications

Hong Kong University of Science and Technology
2018-2024

University of Hong Kong
2018-2024

Schema matching is a central challenge for data integration systems. Inspired by the popularity and success of crowdsourcing platforms, we explore use to reduce uncertainty schema matching. Since platforms are most effective simple questions, assume that each Correspondence Correctness Question (CCQ) asks crowd decide whether given correspondence should exist in correct Furthermore, members may sometimes return incorrect answers with different probabilities. Accuracy rates individual workers...

10.1109/tkde.2018.2881185 article EN IEEE Transactions on Knowledge and Data Engineering 2018-11-13

A Cramér-type moderate deviation theorem quantifies the relative error of tail probability approximation. It provides a criterion whether limiting can be used to estimate under study. Chen, Fang and Shao (2013) obtained general result using Stein's method when was normal distribution. In this paper, theorems are established for nonnormal approximation Stein identity, which is satisfied via exchangeable pair approach coupling. particular, Curie–Weiss model imitative monomer-dimer mean-field model.

10.1214/20-aap1589 article EN The Annals of Applied Probability 2021-02-01

Schema matching is a central challenge for data integration systems. Inspired by the popularity and success of crowdsourcing platforms, we explore use to reduce uncertainty schema matching. Since platforms are most effective simple questions, assume that each Correspondence Correctness Question (CCQ) asks crowd decide whether given correspondence should exist in correct Furthermore, members may sometimes return incorrect answers with different probabilities. Accuracy rates individual workers...

10.48550/arxiv.1809.04017 preprint EN other-oa arXiv (Cornell University) 2018-01-01
Coming Soon ...