AUTHOR=Grelier Guillaume , Casal Miguel A. , Torrente-Patiño Alvaro , Romero Juan 

TITLE=Image sequence sorting algorithm for commercial tasks

JOURNAL=Frontiers in Artificial Intelligence

VOLUME=Volume 7 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2024.1382566

DOI=10.3389/frai.2024.1382566

ISSN=2624-8212

ABSTRACT=Sorting sequences of images plays a pivotal role in augmenting user engagement across a variety of virtual commercial platforms, notably within the real estate sector. Achieving a coherent sequence of images that respects the categorization of specific room types markedly improves the intuitiveness and seamless navigation of potential customers through listings. This study methodically formalizes the aforementioned challenge and broadens its scope to encompass a diverse range of applications by framing it as an ordering problem. The complexity of devising a universally applicable solution lies in the computational demands and impracticality of exhaustive searches for optimal sequencing. Our proposed algorithm navigates this challenge by leveraging a shortest path approach grounded in the semantic similarity between images. It introduces a bespoke solution for the real estate industry, assessing different similarity metrics to adeptly arrange images. Empirical evidence from our dataset underlines the efficacy of this methodology, successfully organizing images in an optimal sequence across 85% of the listings.