As we were looking for project ideas, we wanted something innovative and challenging for us to build. That is where chatPHD came from. Although the name may throw you off (there is no correlation) our product is a simple Ai tool that takes in symptoms you may have and outputs diseases that could be related. It will also provide the user with medications to take as well as a short list of links that may be useful.
We build the model using a dataset we found online using python pandas, sklearn, and several other libraries. We hosted our site on DreamHost and got the front end written using HTML, CSS, and JavaScript. We used Django as our database for the backend.
Although we learned to use all the tools needed for the project, we ran into challenges such as poor quality data sets as the values weren't accurate by definition, causing our model to be skewed. We were also very inexperienced in the developing world and struggled with basic design. These two main challenges caused us to lose sight of our end goal and be forced to settle with what we had.
However, even passed all the difficulties we faced, we are proud to say we learned a lot from this hackathon. We learned how to host a website, set up the file system, and code it using HTML, CSS, and JS. We also learned how to create a model and the steps needed to run the model itself near the end.
All of this makes us feel accomplished, knowing that rather than doing what we know, we pushed for something greater and learned from our experiences. We hope to develop our product over time, adding more features as our skill set grows.
Log in or sign up for Devpost to join the conversation.