Evaluating classifier predictive performance in multi‐class problems with balanced and imbalanced data sets

Abstract

A major issue in classification problems arises when dealing with class imbalance, which requires the adoption of a suitable performance measure able to handle imbalanced data sets. This paper introduces the Balanced 𝐴𝐶 1 and its weighted version Balanced 𝐴𝐶 2 as classifier performance measures suitable for both balanced and imbalanced data sets. The performances of the proposed measures are compared against those of other well-known performance measures through an empirical comparison using several algorithms and data sets.Moreover, the applicability of Balanced 𝐴𝐶 1 is showcased through an illustrative example dealing with steel plate faults classifications, where class imbalance typically occurs due to non-common defects which, though rare, may seriously impact steel quality.

Interpretable Classification of Wiki-Review Streams

“…Furthermore, macro results enable comprehensive evaluation considering the whole set of target classes, while micro-averaging considers the target classes individually. The latter is beneficial in imbalanced classification problems [74], [75], [76]. Finally, run-time is measured to compare the performance of the different models.…”

The ENBIS‐21 quality and reliability engineering international special issue

“…Chuquin et al 4 present a straightforward extension of the K-means clustering to spatially correlated functional data, with an application in the analysis of the normalized difference vegetation index in a large region of Ecuador. From a more general perspective, Vanacore et al 5 address the problem of assessing prediction performance in the multi-class classification of imbalanced data sets. In the optimal Design of Experiments (DoE) Pesce et al 6 offer new results that are oriented to bias reduction in large datasets and protection against confounders.…”

“…present a straightforward extension of the K‐means clustering to spatially correlated functional data, with an application in the analysis of the normalized difference vegetation index in a large region of Ecuador. From a more general perspective, Vanacore et al 5 . address the problem of assessing prediction performance in the multi‐class classification of imbalanced data sets.…”