Students should be aware of some basic concepts in statistics (variance, cross tables, conditional probabilities), management (marketing) and micro-economy. They also should be informed with multivariate descriptive basic algorithms (PCA, linear model) or have ideas on these topics.
At the end of the course, the student should be able to:
– Build a data based predictive strategy, formalize a scoring problem
– Carry out a research relying on some discriminant analysis methods and decision trees.
– Evaluate the performance, control the reliability and accuracy of a score
This course aims at giving students a global contractor’s competence AND basic autonomy to address a scoring issue
Key words : Data Mining – Scoring – Big Data – Machine learning – Data Science
– Introduction to data-based marketing, risk management and predictive techniques
– Introduction to scoring, ROI and simulation for targeted actions
– Discriminant analysis, Decision trees and Scores.
– Use of a statistical software: data management & statistical methods – carry out a data research and reports
– Interpretation of scores efficiency and reliability
– Alternative statistical or computing approaches: neural networks, k-NN, SVM, random forest.
– Introduction to Text Data, NLP, AI and Text Mining
– Loyalty, up-selling, risk (event, loss and premium), appetence, data strategy
