Lab projects

Score aggregation

Global Wine Score

The Global Wine Score was the first project developed by our team. It is a world ranking that evaluates wines with a unique score, providing comprehensive and comparable information for all industry players.

It is an adjusted average score aggregating the main wine critics. It takes into account their rating modes and their respective scales for providing an indicator that minimizes the subjectivity of the experts.

The website has been developed and is already in production used daily by users around the world.

It is recognized by the scientific community as a major scientific project (put forward by the CNRS).

global wine score
Recommender system

Scientific articles recommender

Scorelab collaborates with ICML and NeurIPS, the two major international conferences on Machine Learning,  in the development of an application to recommend scientific articles to participants.

We are working on the recommendation engine of this application. The model is an extension of the collaborative topic regression model.

It is based on modeling both the textual content and the preferences of users on articles. The application is an open source project and a first beta version was released in August 2017.

nips project
Online ratings platform is a “side project” developed from a simple idea. Create a web app to be able to easily rate either a cake contest or a pitch contest, or even the American elections.

We also wanted to take into account the way each evaluator rates by harmonizing ratings. And finally it had to be as easy to use as doodle. This collaborative project, developed on our spare time, has in the end, given birth to, “the doodle of notation”.

noodle project