Shoutout to Prof. Boichuk for this semester's amazing class and introducing to me the importance of being reproducible, accurate and collaborative!
Python program to scrape a sample of tweets associated with 1-800-Flowers.com. This is part of a project in the Marketing class of University of Virginia McIntire School of Commerce's Integrated Core Experience.
Results are used to conduct brand equity analysis about common words people think of when 'Margaritaville' is mentioned.
- For retrieving historical tweets
- nltk
- For tokenizing words and counting frequencies
- nltk.corpus import stopwords
- For removing stopwords (words that add no value to data)
- string
- For removing punctuation
- time
- For counting time taken for each data extraction
- random
- For getting random sampling of dates
- matplotlib.pyplot
- For plotting the frequency distributions
- operator
- For converting nltk dict into tuples
- csv
- For exporting data to csv
- re
- For extracting username from twitter permalink
Program will create a new "tweets.csv" file within current directory