Sparse data to structured imageset transformation

TL;DR

This paper proposes converting sparse datasets into structured images to leverage convolutional neural networks, demonstrating improved classification performance on public datasets by creating visually distinguishable shapes.

Contribution

The novel approach transforms sparse data into images with meaningful structure, enhancing CNN effectiveness for large-scale sparse datasets.

Findings

Improved classification accuracy on two public datasets

Visual structure in images aids CNN learning

Method outperforms traditional sparse data handling techniques

Abstract

Machine learning problems involving sparse datasets may benefit from the use of convolutional neural networks if the numbers of samples and features are very large. Such datasets are increasingly more frequently encountered in a variety of different domains. We convert such datasets to imagesets while attempting to give each image structure that is amenable for use with convolutional neural networks. Experimental results on two publicly available, sparse datasets show that the approach can boost classification performance compared to other methods, which may be attributed to the formation of visually distinguishable shapes on the resultant images.