Dank Dates

Inspiration

As college students, we have all experience poor interactions on dating apps. Ghosting, mismatched interests, dead conversations are far too common. We decided to make Dank Dates to combat this via machine learning analysis of your conversations. Our analysis will provide recommendations and feedback on how to improve your conversations and better your experience.

What it does

First we have each user rate a series of memes on a 'Like' or 'Dislike' basis. We use this to calculate a score that determines your compatibility with other users. Users with high compatibilities are matched and a chat is allowed between them. We then use the google-cloud-language api to analyze their current and previous conversations for topics they are interested in and their current mood. We refresh our results as messages are sent and provide feedback on how to better the conversational experience.

How we built it

Flask and google-cloud-language backend and nodejs, yarn, javascript frontend.

Challenges we ran into

We initially tried to use Docker for cross-system code compatibility, but it did not work out. As we progressed in our code it became difficult to split up the work due to the nature of the code.

Accomplishments that we're proud of

We made a working demo that properly analyzes conversation and provides appropriate feedback.

What we learned

We learned how to use google-cloud-language api, gained experience with nodejs and yarn, how to use flask and mongo.db.

What's next for Dank Dates

Improvements on analysis, changes to suggestions and more suggestions, a more optimized and visually appealing user interface. User information management (logins/passwords).

Built With

• cloud-natural-language
• domain.com
• flask
• javascript
• node.js
• python
• yarn