Analyzing Social Behaviour Using NLP
Poorva Sharma, Shrey Garg · Journal of Emerging Technologies and Innovative Research · 2020
One of the major reasons behind mental ill health in recent times has been depression. Depression also contributes towards the rising number of suicide rates all over the world and also leads to the impairment of daily life. Researchers and doctors have been trying to find out ways to detect depression in people in its early stages itself so that the person can be prevented from the more dangerous outcomes of depression in the later stages such as the danger of committing suicide and harming himself. In recent times social media has come to the rescue by emerging as a platform to annunciate information online. These online networks are used by legions of social media influencers to convey their personal experiences, thoughts and social ideals. Therefore, through this research paper we take a look at the future of social media to forecast, even prior to onset, depression in online personas. People suffering from depression have shown a tendency to communicate via online platforms, where they are not accountable to anyone for what they tweet or post, rather than discussing their problems with their families and friends. Also, previous studies have shown that depression also affects the language usage. In this paper, Naive-Bayes classifier has been used on twitter feeds for conducting emotion analysis focussing on depression. The results have been presented using the primary classification metrics including F1-score, accuracy and confusion matrix.