Inspiration
College students don't ever want to spend money on food, but where can you get food without having to pay for it? We decided to build an intelligent webapp to find out!
What it does
Don't worry, we've got your back; using machine learning and natural language processing, we find all the free pizza (the only food that the people really want) on campus, and present to you in an easy to read map format.
How We built it
Filters emails and tweets to locate free pizza on the RIT campus. We classify whether or not the tweet or email is about free pizza using a Support Vector Machine trained on a bag of words model. We then employ spaCy to extract linguistic features of our text using part-of-speech tagging, tokenization, dependency trees, and, most importantly, named entity recognition.
Challenges we ran into
- Detecting and extracting buildings, rooms, and dates.
- Detecting fake free pizza emails.
- Preprocessing emails, managing the data pipeline, retrieving live data and updating our map in real time. ## Accomplishments that I'm proud of
- Tweet and email parsing for text.
- Predictive classification of free pizza or not
- Extraction of dates, times, and locations.
- Utilization of the Google Maps API.
What I learned
- Spacy and other natural language processing technology.
- Word2Vector.
- More about support vector machines in practice.
- Challenges in text processing.
- Dealing with mbox format and tweepy twitter API.
- Database management lessons. ## What's next for RIT Free Pizza Free pizza rochester, USA ,and the world!!! (Then maybe other food)
Log in or sign up for Devpost to join the conversation.