Next Article in Journal
Optimizing Sensor Data Interpretation via Hybrid Parametric Bootstrapping
Next Article in Special Issue
Invisible CMOS Camera Dazzling for Conducting Adversarial Attacks on Deep Neural Networks
Previous Article in Journal
Vision-Based Localization in Urban Areas for Mobile Robots
Previous Article in Special Issue
A Projective-Geometry-Aware Network for 3D Vertebra Localization in Calibrated Biplanar X-Ray Images
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sensors
Volume 25
Issue 4
10.3390/s25041182
Review Report
sensors-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

HGF-MiLaG: Hierarchical Graph Fusion for Emotion Recognition in Conversation with Mid-Late Gender-Aware Strategy

Sensors 2025, 25(4), 1182; https://doi.org/10.3390/s25041182
by Yihan Wang 1,†, Rongrong Hao 1,†, Ziheng Li 1, Xinhe Kuang 1, Jiacheng Dong 1, Qi Zhang 1, Fengkui Qian 1 and Changzeng Fu 1,2,3,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Sensors 2025, 25(4), 1182; https://doi.org/10.3390/s25041182
Submission received: 8 January 2025 / Revised: 29 January 2025 / Accepted: 12 February 2025 / Published: 14 February 2025
(This article belongs to the Special Issue Deep Learning for Perception and Recognition: Method and Applications)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper presented the HGF-MiLaG model to improve emotion recognition in conversations (ERC). This step is critical for enhancing human-computer interaction by analyzing multimodal inputs (text, audio, and visual). The paper identifies limitations in prior works, such as insufficient modeling of speaker and temporal dependencies and poor utilization of gender information in ERC.

While the overall paper is well-written, there are still some areas that need to be improved.

-While the model uses gender-aware strategies, it does not discuss ethical implications or potential biases introduced by gender-based modeling.

-There is a risk of model overfitting specially with smaller dataset. Authors may show some results to ensure that result is not overfitted.

-The model struggles with class imbalances in the MELD dataset, which could affect real-world applicability

-Comparison with state of the art would add value to the obtained results

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper studies a hierarchical graph fusion for emotion recognition in conversation with mid-late gender-aware strategy. This paper has certain research and application value, but there are some deficiencies need to be improved.

1. Please supplement the reason why the categorical cross-entropy and L2-regularization are used as the loss function in the paper.

2. Please provide additional details on the practical applications of the proposed method and existing methods, as well as the experiment comparison results and analysis.

3. The analysis of the experiment results is rather weak, as well as the description of the experiments.

For the above reasons, we think this paper will be published after revisions in the Sensors.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have revised the shortcomings of the paper, we think this paper can be published in the Sensors.

Back to TopTop