1.Obtain 3000 tweets using $AMZN (AMZN is the ticker symbol for Amazon). A tweet may be categorized in different groups such as original comment, retweet, feeling, sharing of article, Ad, etc. Looking at a few sample tweets, discuss strategies that you may use to categorize tweets into five to seven categories. Read 1st 30 tweets and classify them into categories that you have identified. Report results and provide comments.

2.In light of your response to the first question, discuss some of the challenges one would encounter in creating an algorithm that would accurately classify these tweets into the proposed categories.

3.Develop a frequency distribution of the most frequent word appearing in the tweets, excluding articles, conjunctions and obvious terms (e.g. and, of, the, Amazon etc.). Include your code AND the resulting bar plot

4.Using your list of important words, create a Wordcloud. This wordcloud should be in gray-scale

5.Now experiment with alternative wordclouds, changing color, order, fonts, etc. to create a visualization that you think is more compelling or informative than the initial one.

6.Carry out sentiment analysis of the tweets and summarize key insights.

7.Obtain 3000 tweets each using $AAPL and $GOOGL, and carry out sentiment analysis of the tweets separately for the two companies. Comment on similarity or differences among the three companies.

Note: This assignment is to be solved on Rstudio.

Link to download the Excel File

https://wetransfer.com/downloads/a518909f0ef9920870ce7f9e0a249d9020180402045513/afe8bcf81c0c2e2401c2a500d69f14db20180402045513/825bdd