Determining Clinical Depression From The Analysis of Socio-Economic Attributes

In recent years, clinical depression is increasing among mass people at an alarming rate. In most cases, specially, in under developed and developing countries, people put less importance to the mental health. Moreover, number of study to find pattern among the people having clinical depression is quite less. In our study, we have analyzed some common socio-economic attributes to find the common pattern among the people having clinical depression. After this, we have used this pattern to find whether a person is at a risk of having clinical depression or not. At first, we have analysed the data with basic machine learning algorithms and calculated the performance metrics. Then, we have created an intermediate dataset from the primary dataset which has enhanced all the performance metrics and we have got a maximum accuracy of 92.99%. Finally, we have done some statistical analysis on the attributes lined to clinical depression. Analysing socio-economic attributes to find pattern among the clinically depressed people, generating intermediate dataset to improve the performance of the machine learning algorithms and finding some general observation for clinical depression make this study unique.