This paper concentrates on the primary theme of Tweet Effectiveness in which you have to explain and evaluate its intricate aspects in detail. In addition to this, this paper has been reviewed and purchased by most of the students hence; it has been rated 4.8 points on the scale of 5 points. Besides, the price of this paper starts from £ 40. For more details and full access to the paper, please refer to the site.
This is a critique paper focusing on Xunhu Sun’s thesis “Estimating Effectiveness of Twitter Messages with a Personalized Machine Learning Approach”, which was submitted to the Graduate School of Florida Institute of Technology. In his thesis, Xunhu, focuses his comprehensive research on the measure of tweets by analyzing the total retweets a tweet gets. This he describes as a measure of tweet quality. His thesis proposes a procedure to build a personalized mock-up for a single author, and that uses his past tweets as training data. The study proposes three key types of new features, which are based on tweet content. These features include: Cluster features, Entity, and Pair. The study goes on to combine these features with the features of used in previous works.
As both a social media and news platform, twitter has been studied by numerous researchers interested in examining user’s retweeting behavior. These studies focus on the content that draws an audience to it prompting them to be share and retweet the content in turn spreading it to a wider audience. Xunhu Sun’s thesis outlines an area that has been given less attention. He tries to tackle the question of how to predict the popularity of a tweet. Thus, Xunhu Sun’s methodology estimates how effective an author’s tweet is by examining the features from the author’s earlier tweets, and training a customized model for prediction. His approach introduces three new types of features: Cluster features, Entity and Pair; which significantly surpass earlier features in the aspect of Pearson Correlation Coefficient the prediction on the tested subjects. His work also proposes a modus operandi of assessing tweets in a quantitative approach for each specific author as opposed to taking Boolean predictions (retweeted or not) of posted tweets by prototyping on assorted tweets from a number of authors. In addition, combining Cluster features and Pair, offers the prediction most allied to the authentic number of retweets.