##Description This is a project for CSCE470 (at Texas A&M University) in Spring 2016 which performs sentiment analysis on recent tweets in a specific geographic location. We hope to provide adventurous (but poor) college students a method to choose which city to visit, given current Frontier Airlines deals. These flash deals typically require tickets to be bought that day for flights that are within a certain upcoming date range.
We will:
- Determine the typical difference in time between the sale date and the flight for Frontier promotional emails (optional)
- Collect cities that Frontier Airlines provides promotional deals using a web crawler
- Collect a backlog of tweets to create a overall happiness level for the above cities
- Analyze the sentiments of these tweets
- Compare the happiness levels of these cities
##Resources/Libraries Used
- scra.py
- alchemy API
- openflights API
##How to Run:
- dependencies:
- pip install 'twitter'
- pip install 'tweepy'
- to collect the data:
- You will also need to get your own api key to run this application and insert it in the twit2.py file
- first run python twit2.py making sure that the airport.json file is in the same directory.
- this step should take a little bit of time as it will create the directory data_city then fill in files with incoming tweets
- to run the sentiment on the cities:
- run the line
python alchemyapi.py [API KEY]
- go to the
alchemy_api
directory - run the line
python tweetpy_sa.py
- this should create a file named
<city>_sentiment_results.txt
with the sentiment analysis for the related documents, targeted sentiment analysis for the city, and a weighted score
- run the line