HealthyConvoBot

Inspiration

When we were in highschool, we had a friend that built a twitter bot. This was when Twitter was just becoming popular, at least in our school, so what he did with it was really cool because it would post random rapper quotes, tell you a joke, a lot more stuff that he worked really hard on. So when Twitter revealed their challenge to promote healthy conversations using their API, I thought it would be the perfect time to learn how to make one that would impact others in a similar way.

What it does

We created a small website with some links to promote healthy conversations, such as advice and tips on how to have a good conversation, because in this day and age where conversations have often just become short quips on someone's social media post, it is harder to learn good skills. But the core of the project is the Twitter Bot we made. Around every 10 minutes, it pulls a tweet from the a top trend in the United States and uses Microsoft Azure's cognitive services to run sentiment analysis on it and check whether that tweet was promoting a good conversation. It retweets the tweet, with the score they received and it gets tagged #HealthyConversations

How I built it

The web app was built using React and deployed with Heroku. The Twitter bot used the Twitter Developer API and the Azure Cognitive Services. So this project is 100% node.js, the node It was built with mostly a lot of passion for learning and understanding how things work. At the end of the day, we are all here to just learn something new and support the new technologies.

Challenges I ran into

One of the challenges that the team ran into was learning about the new technologies. Our team did not have any experience with React, the Twitter API, or the Microsoft Sentiment API. It took time for the getting the API's to send data to each other and how to create a react app, but eventually, we were able to create a working model.

Accomplishments that I'm proud of

As a team, we are the proudest of getting a working version of our project up. We were able to achieve the main goal that we had in mind for the Twitter bot from our initial idea. Getting down this made us feel accomplished and gave us a sense of completeness on the project.

What I learned

This was the first hackathon for two of the members of our team, and we definitely learned how to learn new technologies and code under pressure. It was a fun experience and we hope to make even more projects in future hackathons. As for our other teammate, he was able to lead the team and demonstrate how to use these technologies to the rest of us.

What's next for HealthyConvoBot

As the Twitter and Microsoft APIs grow, we hope to make our HealthyConvoBot grow as well. This means better detection for what is considered a Healthy Conversation and having our Bot quote those tweets that are most exemplary. Human conversation evolves rapidly, but we hope to use these APIs to keep recognizing and promoting healthy conversations in our communities.