Technical seminar on Machine Learning algorithms

Aude Doumassi 16 January 2019

The Center for Research and Opinion Polls (CROP) received Dr. Kossi Amouzouvi, a researcher at the African Institute for Mathematical Sciences (AIMS) in Rwanda, on January 5, 2019, as part of our second series of seminars on Machine Learning. Dr. Kossi Amouzouvi’s research interest are in several fields of applied mathematics. One of his areas of interest is the use of Machine Learning algorithms in agriculture and education, particularly the development of algorithms that can better understand the evolution of plantations in order to propose appropriate and timely solutions.

Dr. Kossi Amouzouvi began the presentation with a descriptive analysis of real health insurance data from the United States. A visualization of the data allowed him to highlight the relationship between a couple of variables. After an introduction to Machine Learning he introduced some algorithms of Supervised Learning such as Linear Regression, Linear Discriminate Analysis (LDA), Quadratic Discriminate Analysis (QDA) and others.

With regard to Unsupervised Learning, K-mean, GMM and Principal Component Analysis (PCA) were presented. Dr. Kossi finished the presentation by highlighting the strengths and challenges of Machine Learning in Africa especially in the agricultural field given that the economy of most African states is agricultural. He also gave the example of Rwanda which is investing in the creation of a SMART CITY, project in which qualified people in Machine Learning is strongly solicited. The audience consisted of teachers from the mathematics department of the University of Lomé, the University of Kara, Togolese and Beninese students and professionals.

Rédigé par Aude Doumassi