Adaptive neighborhood rough set model for hybrid data processing: a case study on Parkinson's disease behavioral...

Area of Science:

• Data Science
• Artificial Intelligence
• Machine Learning

Background:

• Hybrid data, combining categorical and numerical types, presents significant knowledge extraction challenges.
• Existing hybrid data processing models require thorough comparison for effective data mining.
• Neighborhood rough sets offer a promising approach for handling uncertainty in hybrid data.

Purpose of the Study:

• To analyze and compare neighborhood rough set-based hybrid data processing models.
• To propose a generic neighborhood rough set-based hybrid model for efficient hybrid data processing.
• To enhance data mining efficacy without discretization, preserving information and practical meaning.

Main Methods:

• Developed a generic neighborhood rough set-based hybrid model for hybrid data.
• Implemented dynamic adaptation of the neighborhood approximation space threshold.
• Utilized a testbed for Parkinson's disease patients to evaluate the scheme.

Main Results:

• The proposed scheme effectively handles both numerical and categorical data adaptively.
• Achieved an accuracy of 95% on the Parkinson's dataset.
• Outperformed existing schemes in adaptive hybrid data processing.

Conclusions:

• The research provides a robust and adaptive solution for hybrid data processing challenges.
• The proposed model enhances data mining efficacy, particularly for complex datasets like those in medical analysis.
• This work advances hybrid data processing techniques, offering improved accuracy and information preservation.