Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jun 29, 2026

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Adaptive Hierarchical Evidence Fusion for Sensitive Field Detection in Structured Data: A Gated Residual Correction

Junpeng Hu^1,2, Xiao Guo², Jinan Shen²

¹School of Cyber Science and Engineering, Sichuan University, Chengdu 610065, China.

Entropy (Basel, Switzerland)

|June 26, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Nontarget Screening and Health Risk of Hazardous Volatile and Semivolatile Organic Compounds in Premade Food Using a Molecular Network Assisted Data Collation Workflow.

Environment & health (Washington, D.C.)·2026

Same author

Weather-Specific Individual Exposure to Organic Contaminants in Fine Particulate Matter by Nontarget Screening and Risk Prioritization.

Environment & health (Washington, D.C.)·2026

Same author

Methyltransferase-like 14 alleviates neuronal ferroptosis in Alzheimer's disease by regulating the peroxiredoxin 6/apoptosis signal-regulating kinase 1 signaling pathway.

Neuroreport·2026

Same author

MI-Net: a dual-stream feature network based on Mamba-convolutional neural network (CNN) fusion to predict intradialytic hypotension using multimodal data.

Quantitative imaging in medicine and surgery·2026

Same author

SRP-Net: sensitive risk propagation network with asymmetric cross-attention for educational data.

Scientific reports·2026

Same author

Intratumoral <i>Enterobacter hormaechei</i> drives gemcitabine resistance in pancreatic cancer via <i>cdd</i> <sub>L</sub>-mediated drug inactivation.

iMeta·2026

This study introduces a novel Hierarchical Gated Residual Network (HGRN) for sensitive field detection in structured data. HGRN enhances privacy compliance by improving cross-domain generalization and mitigating performance degradation.

Area of Science:

Computer Science
Data Privacy
Machine Learning

Background:

Automatic sensitive field detection is crucial for data privacy and governance.
Existing methods struggle with cross-domain generalization due to heterogeneous data and distribution shifts.
Weak semantic signals and distributional heterogeneity in structured data limit current approaches.

Purpose of the Study:

To develop a robust framework for sensitive field detection that overcomes cross-domain generalization challenges.
To improve the accuracy and reliability of sensitive field identification in diverse structured datasets.
To address the limitations of hand-crafted rules and single statistical models.

Main Methods:

Proposed a detection framework using multi-view complementary features and a Hierarchical Gated Residual Network (HGRN).

Keywords:

adaptive feature fusion cross-domain generalization gated residual network sensitive field detection structured data

Related Experiment Videos

Last Updated: Jun 29, 2026

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Constructed a full-spectrum feature system integrating explicit rules and implicit statistical fingerprints (e.g., entropy, character texture).

Implemented a decision mechanism combining a random forest with a learnable gating-and-expert network for residual calibration.

Main Results:

Achieved a Macro-F1 score of 0.9408 on the DeSSI dataset, demonstrating strong in-domain performance.
HGRN mitigated catastrophic performance collapse under frozen-model cross-domain transfer compared to pure neural baselines.
Maintained moderate detection capability in cross-domain scenarios, showing improved generalization.

Conclusions:

The proposed HGRN framework effectively addresses cross-domain generalization challenges in sensitive field detection.
The multi-view feature system and hybrid decision mechanism enhance robustness and accuracy.
Offers interpretable trust allocation between rule-based priors and data-driven correction for privacy compliance in financial and healthcare sectors.