Skip to content

A personal diary that lets you diagnose your feelings using NLP.

Notifications You must be signed in to change notification settings

Naman-Priyadarshi/ForMente

Repository files navigation

ForMente by CitizenFive

build-checks

(The code was initially hosted here - redbrickhacks-2022-submission-citizenfive - but we could not change the visibility of the repository (from private to public) because we don't have the appropriate rights.)

By harnessing the power of Natural Language Processing, ForMente lets you diagnose what you are feeling in the form of a safe and secure personal diary.

Submission for the Red Bricks Hackathon 2022.

Built by Akash Rajoria, Saransh Chopra, Naman Priyadarshi, Paritosh Chaturvedi, and Pratham Chauhan!

The NLP backend API is available here - https://formente.herokuapp.com/ (The API is down as Heroku discontinued its free tier.)

Inspiration

The first person that someone usually talks to about their mental health is themselves. This buries the person with self-doubt that becomes hard to cure. ForMente comes in between to diagnose a person's feelings through Natural Language Processing. It is a personal diary application that lets users know how their daily emotions feel through the text they write.

What it does

ForMente is a personal diary that allows users to write down their minds every day. These texts are stored securely in our database, and users can visit them whenever they want. The main aim of ForMente is to let the user know how they are feeling and that they are not overthinking it. The application runs natural language processing on the texts written by a user and shows them how they are feeling in real-time. The data provided by a user is never accessed by anyone else and is secured safely in a database.

How we built it

We built the application in 4 broad parts -

  • The application

    The application has been built using Flutter and Dart and is available on all Android devices. The UI has been taken utmost care of and is as smooth as it could be. This application is available through GitHub for everyone in the world.

  • The model

    The application uses a model to classify what a user is feeling. This is a Naive Bayes model, trained using scikit-learn and python. We started by training an LSTM neural network, but after spending around 12 hours on it, we couldn't get it to run on an Android device due to technical limitations. After this, we decided to switch to a lighter model, which could be deployed on the free dynos available at Heroku, and we succeeded!

  • The Firebase and Firestore backend

    The Firebase and Firestore backend holds information about a user and their diary entries. This information is kept safe in a remote server and is never shared with anyone. Users can log in or sign up for the application using the Firebase backend and store their diary entries in the Firestore database. The entries are fetched in the application where a user can look back in the past. Additionally, a diary entry also holds information about the emotion that a user was feeling while creating that entry.

  • The NLP backend

    The NLP backend serves as a holder for our model. The backend is written using Python and FastAPI and is deployed on Heroku. The backend API is completely free, independent of the other parts of the project, and easily accessible; hence, developers can use the backend API as a standalone product. The API takes in the text provided by an application and returns an emotion strongly correlated with the text.

Screens

Screens1 Screens2

Video pitch by Saransh

Pitch.for.Formente.by.CitizenFive.mp4

Challenges we ran into

The first challenge we encountered was the repeated failure in pre-processing and fitting the data with the TensorFlow Model. After a sleepless night, we decided to switch to a Naive Bayes model. Developing the Machine Learning backend took most of our time, making it difficult to complete the app. Another big challenge was developing the code for interacting efficiently with the Firebase+Firestore backend. The UI development was also very time-consuming, given that we only had 24 hours.

Accomplishments that we're proud of

Even after a series of hard challenges, we're proud of completing this stamina-focused hackathon. In mere two days, we gained a lot of experience competing in our first offline hackathon. While finding alternatives to the failed approaches, we learned a lot of new technologies that will surely help us in the future.

What we learned

We learned a great deal about Teamwork, Time Management, Organization Skills, and Taking responsibility. We got to explore numerous workshops and talks organized by Ashoka University, which really helped expand our horizons. We were fortunate enough to meet some amazing people from various universities, everyone executing a unique idea.

What's next for ForMente

ForMente is still at stage zero but is still available for and tested for real users. The application would ideally be published on the play store in the future, through which it would be accessible to everyone. Additionally, we aimed to assist people suffering from mental illness, but due to the time limitations, the app only points to their emotional state and does nothing about it. The user can still go through their past notes and see what the app told them about their emotions. This can help the users grow as a person and feel proud of their achievements. In the future, we don't want to keep this project just as a project, but we want it as a product for people who are unhappy with their mental health.

About

A personal diary that lets you diagnose your feelings using NLP.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published