Inspiration

People want to get to information quicker and websites are quickly becoming obsolete. We decided to help Student Loan Company ride the wave of the oncoming conversational chat interface wave.

What it does

Loanly is a collection of services which allows any student to access information using natural language, either on the Amazon Echo or through an automated text bot that understands natural language over text.

How we built it

We used Python to scrape data from the Student Loan Company website, which was then fed into a noSQL database. We also set up a mock database that mimics user info. That way, users can access information about their loan themselves. This noSQL interface was queried using Javascript libraries by the Alexa and Twilio bot to provide structured responses towards user queries.

Challenges we ran into

Getting Alexa to understand accents was difficult. It took us some time to fix this problem. Scraping data from the Student Loan Website was difficult due to bad html and lack of consistent structure. Moreover, translating it into structured data that can be easily queried was tricky too.

Accomplishments that we're proud of

We got all of our features done in time.

What we learned

We learnt a lot of additional Alexa functionality. Web scraping was a new thing to learn too. Integrating multiple with one database.

What's next for Loanly

Improving the database and contextual understanding.

Domain.com - We used a domain.com coupon to get loan-ly.net

Share this project:
×

Updates