As Princeton students, we often found ourselves at a loss for the amount of times we had to access Princeton affiliated websites each day. Often times, it would be such a quick and mundane look at BlackBoard or just glancing at a Piazza post. Inspired, we decided to allow Princeton students to ask questions quickly and seamlessly. Thus, PrincetoNN was created.
PrincetoNN is an unique Alexa skill application that is built for Princeton University. Users can query any Amazon device with Alexa enabled and ask for Princeton specific events, dining options, TigerBook queries, Piazza posts, classes/lectures, and professors.
PrincetoNN was built using Amazon Web Services and the Amazon Developer Alexa Skills Kit. Our application was trained on a large data set of commands to ensure flexibility for users. The backend was built with Node.js 4.3.2. We also used JSON and XML for parsing data from streams.
We had issues with authentication with TigerBook's API. Also, we had some difficulty parsing all the dining hall data since it was extremely long.
We were all very happy that we were able to pick up Amazon Alexa skills and Amazon Web Services (e.g. Lambda) very quickly. Furthermore, two of our members did not know JavaScript before this so it was a great learning experience for them as well!
Above all, we learned how to function as an efficient programming team. Although initially we took a while to make sure we were on the same page with Git, we worked together and helped each other become better programmers.
We plan to train Alexa on a larger training set so that it can respond to a larger array of commands. Furthermore, we aim to take advantage of the other Princeton web feeds so we can deploy a truly useful voice assistant specific to Princeton students.