Rgtsp: a generalized top scoring pairs package for class prediction

Area of Science:

• Bioinformatics
• Computational Biology
• Machine Learning

Background:

• Top scoring pair (TSP) classifiers offer interpretable and robust classification rules.
• Existing TSP methods can be computationally intensive, limiting their application.

Purpose of the Study:

• To present a parallel implementation of the TSP classifier to reduce training time.
• To introduce extensions for enhanced classification performance, including a multi-class approach.

Main Methods:

• Developed a parallel implementation of the TSP classifier utilizing OpenMP libraries.
• Integrated extensions for multi-class classification capabilities.

Main Results:

• The parallel implementation significantly reduces the training time for TSP classifiers.
• The multi-class extension shows potential for improving overall classification accuracy.

Conclusions:

• The parallel TSP classifier offers a computationally efficient and robust method for sample classification.
• Extensions provide enhanced performance and broader applicability in biological data analysis.