RMP Sentiment Analyzer.
We initially discussed how significant it is to select good professors for our courses. As college students, we highlighted how RateMyProfessor is a very useful platform to get to know our professors through multiple student reviews and ratings before actually taking them. We were motivated to utilize the available Google APIs and this drew our attention to natural language processing. We wanted to come up with a unique rating system based on the comments. We believe that words generate a more realistic overall rating than an aggregate of whole ratings. This is what inspired us to explore the use of Google Cloud Natural Language API on the comments posted in RateMyProfessor.
What it does
This application inspects the comments for a given institution or professor and identifies the predominant emotional opinion within the text. It primarily returns the sentiment associated with the comments along with the magnitude of emotion. Based on all the available comments, the application draws an overall sentiment rating.
How we built it
For the backend of the application, we used Python along with Google Cloud Natural Language API. For the frontend of the project, we used Java.
Challenges we ran into
This was our first Hackathon and we were all very excited to participate in this event. We did not have much experience with Hackathons so we were kind of lost at the beginning. Finding proper teammates was an initial challenge and we ended up forming a group in which no one had much experience with front-end-development. We did not even have much experience working with APIs but we took up the challenge and decided to learn as we progressed. We started slow but we picked up our pace as the majority of the work only started on the second day. The only little experience we had with front-end-development was with Java and that was a part of the challenge as well.
Accomplishments that we're proud of
We faced a lot of challenges throughout the process but we overcame it and made it a fun learning opportunity. We did not have experience with what we were focused to work on, but we learned it and we managed to come up with a working result.
What we learned
We learned a lot about the Google Cloud Platform and its APIs. We spent a lot of time dwelling on the platform and it was completely new for us. We also got the opportunity to work on the front-end and the workshops helped and taught us a lot too.
What's next for RMP Sentiment Analyzer
Firstly, definitely a better interface of the application. We can also do a more detailed analysis of the results that we obtained like comparing the RateMyProfessor ratings with our sentiment analyzer ratings. Furthermore, we can also extend it to check for spam or dubious comments.