-
-
The home page of our webite
Inspiration
When one thinks about a lockdown, what do you imagine? Being trapped goodness knows where, until goodness knows when with no means to escape. And there lies the crux of the two-tier problem we are trying to solve. Firstly, we are solving a very common issue faced during the lockdown - boredom - by simulating a lockdown experience in the form of an interactive game. This game is an interesting spin on the word lockdown which is aimed to bring some sparkle into people’s otherwise dull lives. The second, and even more important problem we are trying to solve, is mental illness. This has stemmed due to lack of human interaction, increased boredom and rise in use of social media. In fact, the number of people who reported depression had increased by more than 2 times during the lockdown, compared to the same time period in the previous year.
While this product’s market will be based on statistics, our inspiration was much more personal. The web app is a result of the challenges we faced with boredom during the lockdown, and the lack of psychological aid during these trying times. We watched a close friend of ours struggling with coping up with stress and worry during the lockdown and while he was lucky to be able to reach out to his friends, not everyone will be as fortunate. This inspired us to find a simple solution and this hackathon was the spark we needed to light our fire.
What it does
On the surface of the web application, it appears simple and minimalistic which adds to its charm. This 2d interactive game is sure to engage any audience during the lockdown. The controls and rules are extremely basic so as to not bore the player, which is the exact problem we are trying to solve. The game relies on just four arrow keys to control the sprite while the player attempts to answer simple questions. The light hearted nature of the game is meant to ease and simultaneously engage the player’s.
The backend of our solution is, however, much more complex due to the artificial intelligence involved. As the game progresses, the user will have to answer questions to proceed. The answers to these questions are not black and white, rather, based on the user’s response, the AI will analyse the answers and act as a temporary therapist to the player by providing advice related to mental wellbeing. All questions are verified by real life psychologists to be useful in finding out the thoughts and feelings of a person.
How we built it
To build it, we first analysed the mental problems facing people so that we could create an appropriate survey using which we would be able to analyse the mental health condition of the person. We got most of our information from this website which gave us excellent descriptions of the various mental health problems that people face.
The app itself has two parts:
- The game
- The neural network
We built both the game and the neural network ourselves, the former using javascript and the latter using python. Both of them, being advanced builds, were a real challenge but we managed to overcome the odds to build a full fledged app.
(BTW the neural network didn't like us and decided to stop working so we replaced it with another AI which gives a full diagnosis.)
Challenges we ran into
The first challenge we faced was right at the beginning of the competition where we took a lot of time to decide upon the problem we were trying to solve. This was also a result of a hackathon competition which had just concluded an hour and a half after the start of the competition. Even once we decided to tackle the issue of mental health, we were met with the challenge of having too ambitious an idea which would involve an enormous number of animations, not quite feasible to complete in less than a day. Nevertheless, our team came together and came up with a feasible solution which was also inspired by troubles we ourselves had faced in the past.
Even after coming up with an idea, it wasn’t all smooth sailing. We were met with multiple speed bumps namely in the form of animations and code. We temporarily struggled to make the movement of the on screen sprite look realistic but thanks to our previous experience in flask coding we were successful in making it work. Another issue we faced was with the accuracy of the AI system which was as low as 30%. This is because, us neither being psychologists nor having sufficient training data available to us over the course of the hackathon, were unable to perfectly train our neural network.
Accomplishments that we're proud of
Right from the important problem we tried to solve, to the unique solution and everything we learnt over the past 24 hrs has been an accomplishment in our eyes. We were especially pleased with how we interpreted the problem at hand in a unique and creative way. We were glad to be able to complete our project well in time despite taking up an enormous part thinking of a specific problem and coming up with a viable solution.
Coming to the coding aspect of the product, we were especially impressed with how we managed to bring together so many elements into the web application which we had never worked with before. These included using javascript for game animation, using player input for moving characters and so on. Additionally, we were proud of our progress with Artificial Intelligence which was largely mysterious to us up until we decided to implement it in our solution.
What we learned
Through the course of this 24 hour event we learnt a lot more than we signed up for. Firstly, and very importantly, we learnt to manage our schedule better as we had to manage a fully functional web application over and above our various other commitments. Secondly, we learnt how to work very well as a team and were able to delegate tasks to each team member based on his expertise. Another important skill we learnt was to organise and submit a project in order to sell our idea. We learnt how to make our web app appeal to the judges and users through the course of the competition.
Moving on to the technical aspect of the competition, we were also able to quickly get a hang of flask programming which was integral to our web application. We entered the competition quite confident of our programming skills and were humbled by what more we had to learn in order to reach our ambitious goal. We got the hang of game design and user key input which were the backbone of our solution and had they not worked, we would have a very different result today.
What's next for Isolation
This app has a huge market potential due to the seriousness of the problem of mental health and how perverse but hidden it is in our day to day lives. It seeks to expose this hidden danger in a fun way. Therefore, we believe that our app will be an excellent one for governments to use in order to further the goals of putting an end to mental health.
Members
Member 1: Prakamya Singh, IB YR 1, Section C
Member 2: Praneeth Suresh, IB YR 1, Section D
Member 3: Pratyush Bansal, IB YR 1, Section D
Member 4:Rahul Rajkumar, IB YR 1, Section C
Log in or sign up for Devpost to join the conversation.