Inspiration
This is an app to detect emotions in diaries. This app helps people to understand their emotional score's transition and users can use this app to improve their mental health. In addition to those functions, this app also provides the stress level of the user, and suggest some psychological ways to destress depends on their stress level, which helps users to improve their mental health.
What it does
Letting a user put a daily diary in our app, it detects the emotion scores for different kinds, such as anger, sadness, confidence, fear, and joy in range of 0.0 to 1.0 using IBM Tone analyzer, and then based on those psychological analyses, it provides stress level. After getting to know your own stress level, the app shares some useful information to relieve their stress depending on how badly they have been stressed out.
How we built it
We implemented this app with Androidstudio and our team also used Firebase from GCP, Tone Analyzer from IBM Watson, and MPAndroidchart. We used Firebase to store the data of diaries and emotions, Tone Analyzer to analyze the diary and detect emotions, and MPAndroidchart to make line charts and radar charts.
Challenges we ran into
This is our first time to use NLP API, so it was difficult to decide which API to choose, how to use the API, and what kind of result we should get.
Accomplishments that we're proud of
Although it was really difficult to use the NLP API, we finally succeeded to make a graph of emotions and we are really proud of this accomplishment.
What we learned
We learned how to analyze text and detect emotions with Tone Analyzer, how to store data with FirebaseFirestore, and how to make a graph with MPAndroidchart.
What's next for EmoStack
Improve the UI and usability of the app, and also fix the bug that the graph sometimes does not plot the emotions for some reason.
Built With
- android
- android-studio
- bootstrap
- firebase
- ibm-watson
- java
- mpandroidchart
Log in or sign up for Devpost to join the conversation.