Accounting for your mood record and accurately describe the feelings are hard. We are introducing objective, clear and analytical data to the process. Sentiments analyzation through machines also removes the human biases.
What it does
Identify and differentiate the sentiments through users' journal keeping, propose ways to support and relieve from their sources of stresses.
How we built it
The code was written completely in Swift and we cooperated through GitHub commit and pull under a common project named DubHacks 2019. We integrated GCP Firebase to store, retrieve and read data, authenticate users through their email address and password, set up detailed rules. Furthermore, we have implemented Microsoft Azure APIs for sentiment detection and related image recommendation.
Challenges we ran into
Dysfunctional table view. Merge conflicts. New to the APIs so we accustomed lots of code to better utilize the features. Main thread issue in image suggestion.
Accomplishments that we're proud of
Successfully built the application on mobile devices with beautifully UIs, word extraction and related image suggestions.
What we learned
If you talk with others, you always get opportunities and new perspectives. Helping people to reach out for help is our way to try to give back to the community. Utilizing the APIs significantly cuts down the stresses of building things from the beginning.
What's next for Feel Better
- Connect to bigger organizations with FDA regulations in the United States.
- Expand user bases.
- Only display positive keyword highlights when you are not feeling good
- Send highlights of your diary to trusted contact, if the user permits.