-
-
Daily mood screen
-
Chosen mood screen
-
Goals screen
-
Calendar screen
Group Number: 81
Inspiration
As students with busy schedules, we often forget to take breaks and skip meals when we are busy, and these things are vital for mental health. This also happens for many people who are also overworked and stressed such as people in the workforce, those juggling multiple responsibilities, and those who just can't seem to stop working. We know that a lot of people strongly rely on a calendar to plan their day, so we created an app that blocks off time for you to eat your meals and take breaks. This way, the act of taking a lunch break seems like completing a necessary task, and not like slacking off.
What it does
Breaktime analyzes your calendar events for the day and inserts breaks for meals and rest at appropriate times. Our app also analyzes your calendar events for the week to calculate how much time you spend on work and responsibilities in comparison to fun activities and breaks or self-care. We have three main screens. The home view lets you enter your mood for today and it will give you ideas of quests to complete aka ways of self-care to improve your mood. We monitor the users trends and try and notice patterns to help prevent burnout proactively. The goals screen lets you record when you have completed meals, taken your daily break, and other tasks you would like to track. The calendar screen reads from your calendar and shows you how many events you have for the week as well as how long they are. It uses a machine-learning model to categorize events into work, fun, or self-care and shows you how your days are broken up. You can also schedule your breaks and meals for the day with the click of a button! It finds free time in your schedule to do so.
How we built it
Our app is built using React-Native with Expo. We used Native Base as a frontend framework to help with the design. We used Firebase for the backend to store user data. We created an ML model in mage to analyze calendar events to find out how much time a user is spending on responsibilities versus recreation.
Challenges we ran into
This was one of the first times we worked with machine learning. Mage made it a nice experience as it pointed out ways to clean the data. We also learned about ways to improve the model so we redeployed and used a second model with things we learned from the first one. The first model was 64% accurate but the second model was 84% accurate!
Accomplishments that we're proud of
We are proud of creating our first accurate ML model. We learned a lot about data cleaning, using Mage, and statistics related to models. Two of us have taken an ML class before, but this experience gave us actual experience on how ML is used rather than some of the theory we learned in the class.
What we learned
We learned about how to make a more consistent UI to improve the user experience. We also learned about ML.
What's next for Breaktime
We used our own calendars to train the ML model to simulate real-world data for students. But our audience is for anyone who uses calendars to organize their lives such as health-care workers and CEOs of companies. Thus, we would need more data to train the model. We could also try and personalize the model for each person as they might have specific events that relate to a certain category just for them as some events are like relaxation to certain people, but feel like work to others. We can also do more with the data we get. We can find correlations between completed goals and the user's moods.
Log in or sign up for Devpost to join the conversation.