Students of the 2020/2021 cohort are since the beginning of this week immersed in the fundamentals of Advanced Data Science by Prof. Ndèye Niang Keita of the National Conservatory of Arts and Crafts (CNAM), France.

The main objective of this course by video conference is to introduce students to dealing with big and unlabeled data, how to analyse and to find all the unknown patterns or important features. The course also says which Machine Learning methods to use for this type of data.

“In real life, finding data that has been annotated by human experts is expensive, hard and time-consuming for companies. That is why most of the companies prefer unlabeled data where they don’t need to invest time and resources into humans who will label data,” explained Prof. Ndèye.

• 

• 

• 

For three weeks, she will be introducing the students to some aspects of Big Data analytics and unsupervised learning methods such as Factorial Analysis, Clustering for Quantitative and Qualitative Variables.

Assisted by AIMS Cameroon Tutor Berthine Nyunga Mpinda, Prof. Ndèye is laying more emphasis on popular methods such as Principal Components Analysis (PCA) used for data visualization, dimension reduction or data pre-processing as well as K-means Clustering and Hierarchical Clustering used for Data Clustering. These methods, she says, are so important to students in their career goal because it will help them in analyzing Big Data for different types of variables, reducing dimensionality and also clustering data based on their similarities.

During these three weeks, students will also apply these methods to real life problems using customer segmentation, manufacturing and marketing datasets to support the decision-making process of companies in order to consolidate their position in the market, and also increase the value of their products and services.  

Professor Ndeye Niang Keita is a lecturer in Data Analysis at the National Conservatory of Arts and Crafts (CNAM) in France. She is also a member of the MSDMA research team “Statistical Methods for Data Mining and Learning” at CEDRIC (Center for Research in Computer Science and Communication) and lecturer at the Institut de Statistiques of the University of Paris VI. 

She earned a BSc in Mathematics, MSc. in Applied Mathematics, MSc. in Calculation Methods and Mathematical Models from the University Paul Sabatier in Toulouse. She holds a Doctorate in Statistics from the University of Paris IX Dauphine. 

Prof. Ndèye Niang is a specialist in Data Analysis, Data Mining and Big Data Analysis. After completing her thesis on Multidimensional Methods for Statistical Process Control, she worked on the application of Multiple Array Analysis Methods to Quality Control and on the analysis of Qualitative Variables in Data Mining, in particular on correspondence analysis and several discrimination methods, as well as clustering of variables before extracting association rules in large databases. 

Through several Master’s theses and Doctoral supervision, she collaborates with many companies and research centers for the application of advanced statistical methods to real problems in automotive industries, indoor air quality, management of customer feedback and drug side effects, among others. She is currently working on the development of unsupervised and supervised methods for multi-block data.