Inspiration

The difficulty in accessing mental health care such as high costs, time must be taken during the business hours, and transportation.

What it does

The app at its base is a journaling app, but it can help to asses an individual's psychological state over time. It does this by performing sentiment analysis to asses the emotional connotation of certain words the user is writing. Additionally, it works to find key topics that the user is discussing in their entries by using a text modeling algorithm, tf-dif .

How we built it

javascript

Challenges we ran into

The challenges that we faced revolved around the sentiment analysis and text modeling aspects of the journal. When it came to sentiment analysis the issue was that some words the user is writing could be either positive or negative depending on how they were used. So, how do you train the algorithm to understand the different situations. When it came to text modeling, a topic may be an individual, but the user likely won't be using that indivual's name over and over again. So, how do you train the model recognize the text frequency of pronouns, in this case, and draw that back to the individual initially mentioned in the entry.

Accomplishments that we're proud of

We launched and have an initial user base.

What we learned

Teamwork makes the dream work.

What's next for Uplifit

We plan to continue to work on the journal app to perfect the text modeling and sentiment analysis so that it could be used in clinical practice to help asses trends in a individual's psychological state.

Share this project:
×

Updates