Let's make depression a thing of the past. We're in this together.
* https://www.researchgate.net/publication/257879536_Expressed_music_mood_classification_compared_with_valence_and_arousal_ratings * http://www2.psy.unsw.edu.au/dass/Download%20files/Dass21.pdf
There is no doubt, how Covid has messed up with our lives and it has affected the brain most brutally. Reports suggest that there is almost a 50% surge in depression patients after Covid. Most of the team members of our team had experienced depression in some part of their life and we mutually agreed that the feeling of depression is Brutal. Keeping this in mind we decided to make a web application that determines if the person is depressed and if he is, then provides remedies for the same.
Team Straw hat hackers have come up with a web application with a spectacular User interface. Our web application Uses Machine learning and some questions to determine if the user is depressed and if he is; we provide him with the remedies. We ask the user to take a self-assessment test on our Home page. Our self-assessment test uses the web-cam of the user to determine the mood of the person. The team understands that a person’s face can’t totally decide the feelings inside so we have additional questions set to confirm if the person is sad or happy. If we find that a person is depressed, we provide a list of remedies that he can use to overcome his depression. These remedies include suggesting songs to cheer up the mood, suggesting hobbies to follow, a built in game, and more importantly giving access to a large community of happy and Sad people to talk and express your emotions,
We built the app from scratch beginning with a Flask application. After properly configuring the app we began to work on basic html, css, and javascript. From there we had another portion of our team train a model for a machine learning algorithm that would be able to recognize the users mood. We integrated this model into our app through python in which we then combined with our front-end. From here we added an additional questionnaire to further assess the users mental health. Based on these combined parameters the app will conclude what mood they are in and give useful resources displayed on our final page. We plan to in the future use JS or more python in order to create a chat bot to ask the questions. Last but not least we used github to keep track of our source code.
presentation Slides Link : https://docs.google.com/presentation/d/1fMJ3bplDZYMzt132fgDLdVeIFjVRL5l0JNsCKElYvzg/edit?usp=sharing
We learnt how to make and train the model in the hackathon workshop itself as it was really good and informative. We had teammates from around the world so we faced issues with collaboration but instead of complaining we took it as an advantage and we started to work round the clock and hence we accomplished to make a good web app in less time.
we learnt basics of python, html, CSS, and machine learning apart from that we learnt about various cultures prevalent around the globe. Networking was the best part and the communication between the team throughout the project was really good. We learnt ML face recognition too through the workshop laid by Execute hackathon organizers and that increased our intellectual capabilities of Machine learning.
We had amazing plans for the app but due to time constraints, we could only achieve limited features:- We plan to develop a community of psychologists and happy/Sad people so that people can openly chat with them and seek help. We also plan to keep a song recommender and so that when we know the mood of that person we can suggest him/her a song that will help him go through his phase. We also plan to keep meditation tutorials on our website so that people can be more calm and be strong mentally.