Contextualisation of eCommerce Users

TL;DR

This paper introduces a scalable framework for modeling eCommerce user intent by applying NLP-inspired embeddings to capture contextual relationships and topics within user session journeys, with empirical validation of its consistency and stability.

Contribution

It presents a novel NLP-based approach to model user intent in eCommerce, leveraging session data as documents for contextual embedding, which is a new application in this domain.

Findings

Framework effectively captures user session context.

Embeddings show consistent and stable performance.

Method enhances understanding of user navigation patterns.

Abstract

A scaleable modelling framework for the consumer intent within the setting of e-Commerce is presented. The methodology applies contextualisation through embeddings borrowed from Natural Language Processing. By considering the user session journeys throughough the pages of a website as documents, we capture contextual relationships between pages, as well as the topics of the of user visits. Finally, we empirically study the consistency and the stability of the presented framework.