Sleep is for the Week

'The only thing standing between you and the perfect night's sleep' by Ananya Asthana and Risha Mahendkumar

Your average sleep monitoring app that combines machine learning, data analysis, cognitive wellness and neuroscience. We operate on two types of data: user-input (such as caffeine intake, mood, and access to their Google Calendar) and data from other tracking apps (such as Apple Health, Screen Time and especially Apple Watch data) with a special emphasis on college students' lifestyles.

We care both about bettering your sleep and modifying your sleep to better you. This is why we use information such as your caffeine intake and screen time to suggest ways to better sleep hygiene - including but not limited to better memory consolidation, heightened concentration and reduced irritability - and use information like the kinds of events you have on your Google Calendar and your quality of sleep to tell you what kind of day you might face.

Our Logic:

1. Caffeine Intake- Studies show levels of consumption of caffeine- depending on no caffeine, low caffine or moderate-to-high caffeine- can impact a variety of factors: difficulty falling asleep, difficulty staying asleep, day-time sleepiness, sleep duration, and non-restorative sleep. We want to use user-input and Deep Neural Networks to assess caffeine intake.
2. Screen Time- Daily screen usage right before bed has been shown to decrease sleep quality and increase fitfulness throughout the night. Also, studies have linked increased screen time duration with worse sleep overall. We want to use both user input (from the user's settings and calendar) to analyze both screen time and the time between the last opening of a phone and the first event scheduled for the next day.
3. Stress Levels- A more stressful day can be made easier by increasing sleep, both in quality and in duration. We are trying to access a user's calendar app, and by prompting users to rank each event they have scheduled - as high, medium, low stress and high, low effort - we can provide advice on optimal sleep the night before. For instance, if a user has a low-stress but high-effort day tomorrow, the app will prompt them to sleep relatively early the night before, and perhaps provide resources to avoid screens before bed.
4. Mood- The app will be able to analyze a user's sleep the night before and, based on this data, give prompts and advice describing the possible mood of the user due to these sleep levels. A prior night's sleep of 3 hours, for example, could result in a pop-up message based on user data: "Save the late-night scrolling for a scribe - falling asleep earlier will help your memory consolidation for that test in two days."

Built With

• csv
• html
• json
• python