Prediction of Chronic Insomnia using Machine Learning Techniques

Abstract

The world is changed extremely over the last decade by the power of technology. Consequently, human lives are undergoing multiple changes that have both positive and negative effects on human health. A lot of virtual involvements, lack of physical activity and extreme use of radio-wave devices are leading people into various health-related issues and Insomnia is one of them. The disorder is also known as sleeplessness. This can occur independently or can occur as a result of another problem. This may turn into permanent disease and chronic(long-time) insomnia can seriously damage a human brain. However, the presence of insomnia can be detected by different medical tests according to various internal factors of sleep. But this kind of approach is not only expensive but also time-consuming. Expensive tests and equipment are also not available in many developing countries. To bridge this gap we have decided to build an intelligent model based on a machine learning approach that is able to predict chronic insomnia. For acquiring best results 7 different machine learning classifiers were used where our Logistic regression model outperformed all of them. With an accuracy of 98%, our model can easily classify insomniac and normal people.