Related Experiment Video
Updated: Jun 9, 2025

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
Topic-sentiment analysis of citizen environmental complaints in China: Using a Stacking-BERT model
Junling Liu1, Ruyin Long1, Hong Chen1
1School of Business, Jiangnan University, Wuxi, 214122, China; The Institute for National Security and Green Development, Jiangnan University, Wuxi, 214122, China.
Citizen environmental complaints reveal rising noise, waste, and radiation issues. Sentiment is shifting from negative to neutral, highlighting key regions needing urgent attention for pollution resolution.
Area of Science:
- Environmental Science
- Data Science
- Public Policy
Background:
- Environmental complaints are vital for citizen engagement in regulation and pollution identification.
- Analyzing public environmental concerns provides insights into regulatory effectiveness.
Purpose of the Study:
- To analyze the topics and sentiment of 102,782 environmental complaints from China's e-government platform (2016-2022).
- To identify trends, seasonal patterns, and regional hotspots in environmental complaints.
Main Methods:
- Ensemble machine learning model (Stacking-BERT) applied to analyze complaint texts.
- Quantitative analysis of complaint volume, topics, sentiment, seasonality, and geographic distribution.
Main Results:
- Complaint volume showed an "M-shaped" fluctuation; noise, waste, and radiation complaints increased.
- Sentiment shifted from negative to neutral, indicating a positive trend; significant seasonal patterns were observed.
- Negative emotions dominated (70.41%), with noise, air, and radiation as key topics; high overlap between complaint regions and negative sentiment hotspots.
Conclusions:
- Noise, waste, and radiation are growing environmental concerns.
- Sentiment analysis indicates improving public perception, but significant negative emotions persist.
- Specific regions (Guangdong, Hebei, Shandong, Henan) require targeted interventions for environmental complaint resolution.
More Related Videos
09:16Author Spotlight: Optimization of Processing Technology for Tiebangchui with Zanba Based on CRITIC Combined with Box-Behnken Response Surface Method
Published on: May 12, 2023
08:32Examining Online Syntactic Processing of Spoken Complex Sentences in Chinese Using Dual-Modal Interference Tasks
Published on: September 5, 2019
Related Concept Videos
Stereotype Content Model
Quantifying and Rejecting Outliers: The Grubbs Test
Classification of Signals
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
Survival Tree
Building a Survival Tree
Constructing a...
Labeling Emotion