Learning Tversky Similarity

TL;DR

This paper proposes learning Tversky's ratio model for semantic similarity, demonstrating its effectiveness on image datasets and outperforming existing methods.

Contribution

It introduces a method to learn Tversky similarity measures from data, enhancing semantic comparison of objects like images.

Findings

Tversky similarity can be effectively learned from data.

The approach outperforms existing similarity methods on image datasets.

Learning Tversky measures improves semantic object comparison.

Abstract

In this paper, we advocate Tversky's ratio model as an appropriate basis for computational approaches to semantic similarity, that is, the comparison of objects such as images in a semantically meaningful way. We consider the problem of learning Tversky similarity measures from suitable training data indicating whether two objects tend to be similar or dissimilar. Experimentally, we evaluate our approach to similarity learning on two image datasets, showing that is performs very well compared to existing methods.