Fair and trustworthy recommendations in e-commerce using AI

Recommendation systems often favor dominant sellers and ignore user context, leading to biased and irrelevant suggestions. This thesis examines how AI can improve both fairness and trustworthiness in e-commerce recommendations. You’ll analyze existing methods, test algorithms on real-world data, and design a solution that balances market visibility with user relevance.

Recommendation systems are a cornerstone of modern e-commerce platforms. They shape customer behavior, influence satisfaction and directly affect market dynamics. However, the fairness and reliability of these systems are questionable, as studies have shown that many popular algorithms tend to favor products from dominant sellers, creating a biased product offering. This undermines the visibility of smaller vendors and reduces consumer trust.

Additionally, these systems often lack contextual awareness. They may recommend items that have already been purchased or fail to distinguish between multiple users sharing the same account (e.g., family members), leading to irrelevant suggestions. These inefficiencies not only diminish the user experience but also further question the effectiveness of the recommendation engine.

The Assignment

The aim of this thesis is to contribute to the development of fair and trustworthy recommendation algorithms for e-commerce platforms using AI. This assignment addresses both aspects through three main stages:

1. Literature review and analysis

   Explore existing research on bias in recommendation systems. Identify fairness metrics used in literature and industry, and review approaches to mitigate bias. Also, study methods for incorporating user feedback directly into the recommendation process (customer-in-the-loop systems).

2. Algorithmic evaluation

   Compare algorithms that explicitly aim to optimize fairness. Evaluate these algorithms using publicly available datasets (e.g., Amazon Reviews, MovieLens), comparing their performance in terms of recommendation quality, fairness, and diversity.

3. Solution design

   Develop an approach that addresses and combines both fairness and trustworthiness, improving the overall usability of e-commerce recommendations. Measure the impact on the recommendation system’s performance using clearly defined metrics, for example through a prototype webshop.

This thesis should contribute to the academic discourse on fairness in AI, while offering practical insights for improving recommendation systems in online commerce.

About Info Support

Info Support specializes in custom software, data/AI solutions, management, and training and is active in the Finance, Industry, Agriculture, Food & Retail, Mobility & Public, and Healthcare sectors. We provide solid and innovative solutions for complex and critical software issues. Our headquarters are located in Veenendaal (NL) and Mechelen (BE). At present, approximately 500 employees are employed by Info Support.

Info Support’s working method is characterized by a number of core values: solidity, integrity, craftsmanship, and passion. These core values are intertwined in our work and the way we interact with each other.

To ensure that all employees are always up to date with the latest developments, Info Support has an in-house IT Academy that eagerly satisfies the hunger for more or different knowledge and skills.

B2 language proficiency in Dutch is required.