E-Sports Competition Analysis Based on Intelligent Analysis System.

Area of Science:

• Computational intelligence
• Neuroscience
• Sports analytics

Background:

• E-sports analysis requires sophisticated methods for interpreting complex data.
• Existing multi-classification techniques have limitations in accuracy and efficiency.
• Motor imagery electroencephalogram (EEG) signals offer potential for objective performance analysis.

Purpose of the Study:

• To enhance the analytical capabilities for e-sports competitions.
• To investigate and compare various multi-classification methods for EEG signal analysis.
• To develop an intelligent analysis system for e-sports data.

Main Methods:

• A novel two-layer Support Vector Machine (SVM) classifier was designed.
• The classifier was specifically structured for four types of motor imagery EEG signals.
• The system's performance was evaluated using e-sports competition datasets and compared against the Directed Acyclic Graph Support Vector Machine (DAG-SVM) method.

Main Results:

• The designed two-layer SVM classifier achieved high classification accuracy for motor imagery EEG signals.
• The proposed intelligent analysis system demonstrated a significant improvement in e-sports competition analysis.
• The system showed promising results in e-sports competition prediction.

Conclusions:

• The developed two-layer SVM classifier offers an effective approach for analyzing e-sports competition data.
• The intelligent analysis system provides a robust framework for enhancing e-sports performance insights and prediction.
• This research contributes to the advancement of computational intelligence in sports analytics.