Text Analytics Project

Big Data & Analytics

Fall 2020

This is a Text Analysis project from my Big Data & Analytics class focusing on Natural Language Processing and Sentiment Analysis. Given the entire text of Tarzan, my partner and I cleaned the texts and analyzed it using the programming language R, looking for specific pieces of information. Cleaning the text included filtering out stop words, removing punctuation, and removing sparse words. The image on the right shows a wordcloud of all the most common words in the book, a helpful visualization of the data. We found the longest sentences and words in the entire book and marked the parts of speech for the words in those sentences. Next, we analyzed the frequency distribution of the words in the longest sentences. Lastly, we used the SentimentAnalysis package to analyze the sentiment of each chapter, which we then plotted. This project gave me an understanding of text analysis and provided a valuable foundation in data science.