Social media has become a fixture of modern life, a constant stream of information coming and going. Researchers at the University of Pittsburgh examined social media use and sleep in a group of young adults, and found that heavier users of social media are significantly more likely to experience disturbances to their sleep. This research inspired us to build a product that will alert users about their habits.
What it does
The idea is to provide this application as a feature in social networking site through which a user can get information about the sleeping habits and whether it can cause sleeping disorder in the long term. The app will provide a detailed analysis about the sleep and if a user is suffering from insomnia In the first place it will recommend programs such as Cognitive Behavioral Therapy.
For proof of concept, we are analyzing Facebook data. We came up with several metrics that will tell us about the sleeping habits of a user:
Social media volume was a measurement of the amount of time spent engaged daily.
Social media frequency was a measurement of the number of visits to social media sites over the course of a week.
Based on the above metrics, we calculated a Sleep Score that tells us whether the sleeping habits are good or bad (detection of sleep deprivation) and the we send a text message to the user to improve their sleeping habits with a link to a application to see their detailed analysis, and based on that score we provides suggestion for different therapies such as Cognitive Behavioral Therapy (prevention).
How I built it
We used docker to spin up the Mongo DB instance. From end is built using Angular JS, HTML, Bootstrap. We used Java to build the REST APIs and as the underlying language to design the whole application. We used machine learning to predict the number of sleeping hours.
Challenges I ran into
Packaging the application for deployment on AWS.
To get last logged-in time of the user.
Integrating routers in AngularJS.
Accomplishments that I'm proud of
What's next for DSDP
We aim to integrate machine learning models to predict the future sleeping disorders and based on that provide suggestions to ameliorate them.