ProVe -- Self-supervised pipeline for automated product replacement and cold-starting based on neural language models

TL;DR

ProVe is a self-supervised pipeline leveraging neural language models to automate product replacement, manage new product introductions, and enhance demand forecasting in retail, addressing human limitations and operational challenges.

Contribution

The paper introduces a novel self-supervised pipeline that uses NLP models for product replacement, new product categorization, and improved demand forecasting with minimal transactional data.

Findings

Effective product replacement recommendations.

Successful categorization of new products with minimal data.

Improved demand prediction accuracy.

Abstract

In retail vertical industries, businesses are dealing with human limitation of quickly understanding and adapting to new purchasing behaviors. Moreover, retail businesses need to overcome the human limitation of properly managing a massive selection of products/brands/categories. These limitations lead to deficiencies from both commercial (e.g. loss of sales, decrease in customer satisfaction) and operational perspective (e.g. out-of-stock, over-stock). In this paper, we propose a pipeline approach based on Natural Language Understanding, for recommending the most suitable replacements for products that are out-of-stock. Moreover, we will propose a solution for managing products that were newly introduced in a retailer's portfolio with almost no transactional history. This solution will help businesses: automatically assign the new products to the right category; recommend complementary products for cross-sell from day 1; perform sales predictions even with almost no transactional history. Finally, the vector space model resulted by applying the pipeline presented in this paper is directly used as semantic information in deep learning-based demand forecasting solutions, leading to more accurate predictions. The whole research and experimentation process have been done using real-life private transactional data, however the source code is available on https://github.com/Lummetry/ProVe