# Combination of linear classifiers using score function -- analysis of possible combination strategies

## TL;DR

This paper investigates various strategies for combining linear classifiers using score functions, comparing their effectiveness through experiments with heterogeneous ensembles and multiple quality metrics.

## Contribution

It introduces and evaluates four combination strategies for linear classifiers based on score functions, providing insights into their relative performance.

## Key findings

- Simple average and trimmed average are the most effective combination strategies.
- The proposed methods outperform majority voting and model averaging in several quality criteria.
- Experimental results validate the effectiveness of geometrical combination strategies.

## Abstract

In this work, we addressed the issue of combining linear classifiers using their score functions. The value of the scoring function depends on the distance from the decision boundary. Two score functions have been tested and four different combination strategies were investigated. During the experimental study, the proposed approach was applied to the heterogeneous ensemble and it was compared to two reference methods -- majority voting and model averaging respectively. The comparison was made in terms of seven different quality criteria. The result shows that combination strategies based on simple average, and trimmed average are the best combination strategies of the geometrical combination.