In this project, we have chosen to explore some advanced machine learning topics such as those of class imbalance, cost-sensitive learning and explainable machine learning. As our dataset, we have chosen to analyze the African Country Recession Dataset available from Kaggle.

Outcomes derived from this dataset can be used as an indicator to determine whether or not a country is in an economic recession which may be useful for use cases where an investment firm or organization may want to invest in businesses based in a developing country or in the case of investing in properties and a general indicator of economic health is required beforehand. It is also beneficial in the case of personal investment such as deciding to find a job in said country. Finally, it is an interesting indicator as to whether it would be a good holiday destination as countries in recession tend to have devalued currencies making tourists from more economically powerful countries more attracted to visit them due to the favorable exchange rates.

Joint work with Athanasios Kefalas and Petros Tzallas