Inspiration

We have all experienced times when we don't know what to do with our free time, or at least we know we could invest it in something more educational. Our idea is to fill these free times by providing users with summarized, interesting and customized 5 minute readings.

What it does

Use your free time to learn new concepts everyday. FyveBot will scrape the web for subjects you are interested in, combine speech-to-text recognition technology with a smart summarizer to provide the reader with concise, 5 minute reads.

How we built it

We separated the project into FrontEnd, BackEnd and Algorithm Design in order to best utilize our team's resources. The FrontEnd team developed an easy to use interface that would guide the user in order to provide an optimal experience of the platform. The BackEnd team ensured that all the API calls, packages and other external open source softwares were properly linked to summarize and return the material; programming languages used included javascript and python.

Challenges we ran into

Being our first Hackathon, we ran into multiple challenges. Our goal was to develop something autonomous that would benefit the user, however, none of us had previous experience using Microsoft Bots, summarizers or auto punctuations. These were the root to most of our challenges, we spent our initial hours learning how to use these technologies and properly combine them to achieve our goal.

Accomplishments that we're proud of

We are amazed and proud that we were able to actually develop a minimum viable product for our project. We began this Hackathon without deep understanding of most of the skills that were required for the project. After long hours of reading and talking to representatives on campus, we can say we are comfortable using Bots, web scrapers and summarizers.

What we learned

Everyone on the team learned a lot from this project. None of us had ever worked with azure or the Microsoft bot framework, and we got a much better understanding of the capabilities of both. We also learned a bit about deep learning and it’s potential, with the auto-punctuation tool we used. We got more familiar with web scraping, when we used the Bing API and (temporarily) Selenium.

What's next for FyveBot

For the near future, we aspire to get FyveBot proper hosting were we can combine all the technologies required as well as connecting it with Microsoft's authentication services in order to provide a more customized learning experience for the user


For our project we created a GitHub organization with different repos, the main product can be found at hackillinois2k18.github.io

Built With

Share this project:
×

Updates