EmoTex - Detecting emotions for a personalized chatbot with a novel approach

Abstract

Dealing with emotions can sometimes be difficult how tough a person can be. Emotions that specially affect your mental health is one of the most concerning things in the current time. Since some of these even can lead to depression or even suicidal thoughts. Even when it comes to coping with this emotion only a few percentages of people really do go out and get help from others. In other case people just suppress these emotions and go on with their lives without coping them which makes them even worse and as mentioned which lead to bad ways. One of the major reasons is that people don’t like to talk about their emotion and they are either extremely shy about sharing those emotions or embarrassed to express their emotions which they think that make them weak or vulnerable. In the current state there are various numbers of machine learning models that have been created to detect emotion, most of the research that has been conducted are based on Sentiment Analysis based emotion detection and with that said there is no application that out there which is essentially a chatbot that can talk to a person and help with their emotions. When it comes to emotion detection in the research field there are multiple ways of detecting emotion whether it being text base, voice based or image base. In this research that is being conducted the focus is on text-based emotion detection. The solution that EmoTex tries to provide is a chatbot that can detect emotion through text and response to that text according to the emotion that was detected. With the help of the emotion based dataset and the machine learning model. EmoTex will be able to provide a chatbot that a person has a conversation about their emotion and get some help to cope up with those emotions.