A consumer emotion analysis system based on support vector regression model

Area of Science:

• Consumer Behavior Analysis
• Computational Linguistics
• Machine Learning Applications

Background:

• Consumer purchasing decisions are influenced by varied habits and experiences.
• Online platforms require methods to understand diverse consumer preferences.
• Analyzing online comments offers insights into product characteristics and sentiment.

Purpose of the Study:

• To develop a consumer sentiment analysis system for online fitness platforms.
• To predict consumer willingness to pay based on sentiment analysis.
• To enhance online fitness platforms' service quality and marketing strategies.

Main Methods:

• Utilizing an optimized Support Vector Regression (SVR) model.
• Employing Region Convolutional Neural Network (RCNN) for feature extraction.
• Training the SVR model with extracted consumer sentiment features.

Main Results:

• The RCNN-optimized SVR model demonstrated improved accuracy in sentiment analysis.
• Accurate sentiment analysis aids businesses in targeted promotion and sales increase.
• Enhanced analysis helps users identify preferred fitness projects efficiently.

Conclusions:

• The developed consumer sentiment analysis system holds significant practical value.
• Improved sentiment analysis benefits both businesses and consumers in the online fitness market.
• This approach provides actionable insights for optimizing online fitness offerings.