Predictive analytics for economic crisis triggered depression risk level identification among some adults in Nigeria

Abstract

The striking increase in the number of depression cases globally is alarming and can be attributed to a lot of factors such as economic crisis, insecurity, high rate of terrorist attacks, and increase in the frequency of occurrence of natural disasters and epidemics. According to a recent report from the World Health Organisation, more than 322 million people are dealing with depression cases globally. In Nigeria, economic crisis and a high level of insecurity can be regarded as the major triggers for the recent surge in the number of depression cases in the country. A report in 2018 from the study conducted by the World Bank shows that 22 percent of Nigerians, on the average, are chronically depressed. In this research, we explored the use of data analytics in identifying and predicting depression risk levels among government employees in Nigeria. Specifically, we adapted the Beck Depression Inventory-II (BDI-II) psychometric tool for data collection from 500 government employees in Ondo State Nigeria and used a multiclass naíve Bayes (NB) classifier to identify the depression risk levels among the respondents. Evaluation results on discovering depression risk levels by our proposed model in our study area show accuracy and precision of 96% and 92%respectively.

Keywords

• Depression
• Predictive analytics
• Data mining
• Machine learning
• BDI-II Tool
• Economic crisis