An interactive journal experience that provides insight to and engages human emotions

• General info
• Intro Video
• Technologies
• Setup
• Features
• Status
• Inspiration
• Contact
• License

Sensi is a single page application that allows users to create a profile and track their emotional patterns. The user creates a descriptive entry of their mood or day's events, which is processed by sentiment analysis technology and rendered on the user's profile as a color indicative of mood polarity. A chart is included to count and compare mood polarities. Further, a user can balance their current emotional state in a View room of relaxing VR videos, but to access it the user must pass a Smile test! A snapshot is taken of the user every six seconds until their smile is scored at 95% or above.

Sensi on YouTube

• Vue.js
• Node with Express
• PostgreSQL
• AYLIEN Sentiment Analysis
• Microsoft Cognitive Services - Face API
• Firebase
• Heroku

To run this project, install it locally by cloning the GitHub repository.

Front end repo can be found at https://github.com/catherine-o/sensi-frontend

API key/ID for AYLIEN will need to be switched out after 11/16/19. You may sign up for a free trial and replace the key and ID. These are found at the top of api/users.js You will also need to set up your own local database connection as it is currently connected to Postgres via Heroku.

From inside the project directory:

npm install

npm start

Open your browser and go to http://localhost:8080.

router.post('/users/signup', (req, res) => {
    let userExists = User.query().where('username', req.body.username).first()
    .then(function (userExists) {
        if (!userExists) {
        bcrypt.hash(req.body.password, saltRounds, function(err, hash) {
            User.query().insert({
                username: req.body.username,
                name: req.body.name,
                password: hash
            })
            .then(user => {
                jwt.sign({user}, process.env.SECRET_KEY, (err, token) => {
                    user.password = null
                    res.json({
                        user,
                        token
                    })
                })
            })
        })
        } else {
            res.json('Username already taken')
        }
    })
})

router.get('/users/:id', (req, res) => {
    let id = parseInt(req.params.id)
    jwt.verify(req.token, process.env.SECRET_KEY, (err, authData) => {
        if(err) {
            res.sendStatus(403)
        } else {
            User.query()
                .where('id', id)
                .eager('posts')
                .then(user => {
                    res.json({
                        user
                    })
                })
        }
    })
})

• Create user account
• Submit new journal entry for sentiment analysis
• Review previous entries and track emotional patterns over time
• View total counts of each mood polarity in Chart view
• Access VR View room via camera snapshots and smile analysis
• Balance current emotional state by viewing relaxing VR videos

To-do list:

• Build another chart type in Chart view
• Implement a filter function so users may choose VR mood
• Allow users to delete entries
• Complete responsive design functionality

Project is: deployed for desktop browsers at https://sensi-journ.firebaseapp.com. However, will need new API keys for AYLIEN & Microsoft Cognitive Services (Limited Trials)

The inspiration for Sensi came from an interest in AI and Machine Learning technologies. I wanted to build a product that would increase user engagement, such as camera hardware or possibly a VR headset. I was also inspired by futuristic and clean design concepts, such as seen in some episodes of a popular Netflix technology TV series :)

Created by Catherine O'Hara

Feel free to contact me!

GPL v3.0