Inspiration
In modern society, gadgets like the phone are very commonly used. Me and my friends used phones every day. Whenever I turn on the phone, I am always worried that I would get phone addiction. I thought it would be helpful if I knew my level of phone addiction, so I would know for sure when I am getting more addicted.
What it does
The phone addiction predictor first asks for the following information: the user's daily phone use in the week, average weekend phone use, social media hours, and the amount of time spent gaming. Using the information, the predictor predicts the phone addiction level, and shows the percentile of the user.
How we built it
I trained a model using a dataset from Kaggle, created the forms using HTML. and made the webpage look nicer using CSS.
Challenges we ran into
The challenges I ran into included: ensuring that the model was accurate, and correctly displaying a matplotlib chart into the webpage.
Accomplishments that we're proud of
I am proud of displaying a matplotlib chart, and the predictions of the model into the webpage.
What we learned
Things that I learned include: how to correctly display a matplotlib chart into a webpage, train a model and show its predictions in a webpage, and how to caluclate percentile. After spending a lot of time trying to make the model more accurate, I learned that sometimes a model may show less accuracy because the data may not have much correlation. In the project, I wanted to calculate percentile to show the person how they were compared to the rest of the teenagers. The percentile was the percentage of scores lower than the user's score, so I figured out that it was (numer of scores lower than user's) / (the total amount of scores) * 100.
What's next for Phone Addiction predicter
Next steps for the Phone addiction predictor include: recommendations on how to reduce phone use, and daily phone addiction tracker.
Log in or sign up for Devpost to join the conversation.