The objective of this work is to identify practitioner profiles based on positive or negative qualifiers, i.e., to assign a profile to each cluster in the data and to estimate the strength of these profiles, i.e., the number of clusters or profiles that are most representative of the data.

For the execution of the project it is necessary to install the following modules:

• NumPy NumPy Documentation

  pip install numpy

• Scikit-learn Scikit-learn Documentation

  pip install -U scikit-learn

• MatPlotLib MatPlotLib Documentation

  pip install matplotlib

• Panda Panda Documentation

  pip install pandas

• Factoshiny Factoshiny Documentation

  install.packages("Factoshiny")

• FactoMineR FactoMineR Documentation

  install.packages("FactoMineR")

The data Analysis and Classification folders contain 2 files: a file that creates python functions and methods and a test file that uses its functions to answer our problem.we find all our data in the file, the project report in reports. The clustering file shows the codes used in R for the clustering part of the project. Our test files allow us to reproduce the results obtained in the project.

Projet made by Congo Job, Master student at CSMI in Strasbourg under the supervision of SCHNITZLER Christophe , Christophe Prud'homme and AGHILI Joubine