-
an example of a chat between user and chatbot
Inspiration
We were inspired to do this because we had all experienced the struggle of trying to stay up-to-date with current legislation using google searches or news outlets due to complicated legal/political language.
What it does
jargone is a conversational UI that allows its user to ask a question about current updates on a specific category of legislation. It then asks the user to choose which resolution/bill they want to read about, and once they have selected an option, the legislation is simplified by cutting out/simplifying certain vocabulary so that the user can understand it better.
How we built it
We used Microsoft Azure to create the chat bot that interacts with the user. The congress API was used to get information about legislatures once the user entered a key word/intent such as "guns" or "healthcare". The API returned us information such as the summary of the bill/resolution, the policy maker in charge, and the political party, which we reported back to the user through the chat bot. Each of the results had a button called "simplify", which used our simplification code/method. This method replaced what we deemed "hard words" with more common language. The simplified text was returned to the user through the chat bot.
Challenges we ran into
We tried multiple resources for creating the chat bot, before finding one that allowed us to access other API's and personalize the actions. We also had trouble when first attempting to use the Congress API, as it had little documentation and was harder to use compared to other API's.
Accomplishments that we're proud of
We have really improved as a team from our last HackTJ, and are excited to show off our project! We tried new things, such as a chat bot, and interacted with many companies.
What we learned
We learned many things such as implementing an API, HTTP clients, and creating a conversational UI, specifically working with Azure. We also improved on our previous knowledge through writing methods such as "simplify" that required more extensive algorithms and data structures (such as hashmaps).
What's next for jargone
We want to continue to improve our simplification methods so that the essence of the legislation can be captured and be more easily understood. We would also like to improve intents to field a larger variety of questions from users.
Log in or sign up for Devpost to join the conversation.