Generating in-store customer journeys from scratch with GPT architectures

We propose a method that can generate customer trajectories and purchasing behaviors in retail stores simultaneously using Transformer-based deep learning structure. Utilizing customer trajectory data, layout diagrams, and retail scanner data obtained from a retail store, we trained a GPT-2 architec...

Full description

Saved in:
Bibliographic Details
Published inThe European physical journal. B, Condensed matter physics Vol. 97; no. 9
Main Authors Horikomi, Taizo, Mizuno, Takayuki
Format Journal Article
LanguageEnglish
Published Berlin/Heidelberg Springer Berlin Heidelberg 01.09.2024
Springer Nature B.V
Subjects
Complex Systems
Condensed Matter Physics
Customers
Fluid- and Aerodynamics
Machine learning
Physics
Physics and Astronomy
Recent Advances in Complex Systems
Regular Article - Statistical and Nonlinear Physics
Retail stores
Solid State Physics
Online AccessGet full text

Cover

More Information
Summary:We propose a method that can generate customer trajectories and purchasing behaviors in retail stores simultaneously using Transformer-based deep learning structure. Utilizing customer trajectory data, layout diagrams, and retail scanner data obtained from a retail store, we trained a GPT-2 architecture from scratch to generate indoor trajectories and purchase actions. Additionally, we explored the effectiveness of fine-tuning the pre-trained model with data from another store. Results demonstrate that our method reproduces in-store trajectories and purchase behaviors more accurately than LSTM and SVM models, with fine-tuning significantly reducing the required training data. Graphic abstract
ISSN:1434-6028
1434-6036
DOI:10.1140/epjb/s10051-024-00778-1