We were inspired to bring this idea to life when a person close to us took his own life because he was not able to get help he needed. We all experienced the struggle of having to endure the loss of a loved one. At Aeterna, we believe that no one should have to go through such a tragic catastrophe. In response, we created Aeterna to answer others' cries for help before they manifest into something worse.
What it does
Aeterna means eternal in Latin. In our domain it is aeternahelp.com, meaning it is eternal help. Essentially, Aeterna intelligently analyzes the social media presences of specified profiles to identify and expose dangerous behavior or dialogue indicative of potentially dangerous activity. It goes as far as to give a decent guess of the risk percentage of a user, as well as a history of how their mood has developed and the most worrying posts by a user. Using Aeterna, a person's warning signs will not go unnoticed.
How we built it
As far as intelligent analysis goes, we utilized IBM's Watson's "sentiment analysis" feature to read the posts of social media accounts and determine the negativity of their content. We also used Twitter's API to scrape data from peoples' posts. We wanted to also scrape from Instagram, Tumblr, and Facebook, but did not get the time. We will implement them in the future.
Challenges we ran into
Our biggest issues in the development of our project were concurrency/compatibility issues with the work done by our teammates before they left us. As a result, we had to start over and redo their work from scratch, which proved extremely detrimental to our process. We also had a few issues with vectorization of our graphics for UI design.
Accomplishments that we're proud of
We are very proud of our learning how to use Watson on the spot and for such a meaningful cause. We are also proud of our sleek/material UI for the homepage and color choices.
What we learned
We learned a lot about how team communication should be made more clear before team members leave such that others can continue the work they began, as well as how IBM APIs are used in general for machine learning-based applications. As with every hackathon, we learned a lot about time management.
What's next for Aeterna
We will further be developing Aeterna to scrape and analyze more social media outlets, including photos from Instagram that Watson can analyze as well. Hopefully, this will be used in the future to help those in need and make their cries for help be heard. Perhaps it could even save a life one day.