Recommender systems are on most e-commerce websites, helping customers to find the best product or service. In this thesis, we focus on interactive recommenders (IR) which are used to help customers with complex products by asking them simple questions about their needs. Those recommenders can be combined with event driven architecture (EDA) that saves each state of the system lifetime. This combination allows to generate much more useful data and to be able to analyze it more effectively. Recommender combined with EDA are very powerful and generate lots of data about the user’s needs but still very little is known about how to best take advantage of these data. After defining which data is relevant to the business, we have built a proof of concept to show how and what data can be extracted from these systems. This proof-of-concept extract data from an EDA-based IR to show a new way to benefit fully from IRS-generated data.
| la date de réponse | 31 août 2021 |
|---|
| langue originale | Anglais |
|---|
| L'institution diplômante | |
|---|
| Superviseur | Patrick Heymans (Promoteur) |
|---|
Business insights from interactive recommendation systems: a proof of concept
Albrecq, J. (Auteur). 31 août 2021
Student thesis: Master types › Master en sciences informatiques à finalité spécialisée en data science