Analysing tweets in R programming

1. Use rtweet library to download 1000 tweets that the company posted. Save these tweets as “[login to view URL]”. 2. Use rtweet library to download 1000 tweets about the company you selected. Save these tweets as “[login to view URL]". 3. Examine the source column of both the company and the public tweets to see the source of tweets. Find out how many different levels of sources exist in the public and company tweets. 4. Draw a bar plot of the top 10 most frequent tweet sources for both company tweets and the public tweets. Label each bar with the source name. 5. Comment on your bar plots. 6. By using an appropriate statistical test, test whether retweeting is independent of the tweet source that the public posted. Use the “source” and “is_retweet” columns to get the source and retweet information. Group the sources as; “Salesforce - Social Studio”, "Twitter for Android", “Twitter for Ipad”, “Twitter for iPhone”, “Twitter Web App”, “Twitter Web Client” and “Other”. 7. What is the conclusion of the test? Interpret your results. 8. Calculate a 95% confidence interval of the text width used in the tweets that the company posted. Use the “display_text_width” column to get this information. 9. Combine [login to view URL] and [login to view URL] and save as tweets. 10. Clean and pre-process the data (use TFIDF weights in your analysis). 11. Compute the most appropriate number of clusters using the elbow method for the combined tweets by using cosine distance. 12. Cluster the tweets using the most appropriate clustering method. 13. Visualize your clustering in 2-dimensional vector space. Show each cluster in a different colour and the tweets in [login to view URL] and [login to view URL] with different symbols in your visualization. 14. Comment on your visualization. 15. Compute the proportion of [login to view URL] at each cluster. Print these proportions. 16. Which clusters are dominated by the public and which are dominated by the company? 17. Draw a word cloud and a dendrogram of these two clusters to understand the theme of the clusters. 18. Find the most popular 10 friends of the chosen Twitter handle. 19. Obtain a 1.5-degree egocentric graph centred at the chosen Twitter handle and plot the graph. The egocentric graph should contain the most popular 10 friends of the chosen Twitter handle. 20. Compute the betweenness centrality score for each Twitter handle in our graph. List the top 3 most central people in your graph according to the betweenness centrality. 21. Comment on your results.