Instruction

You are required to conduct some sentiment analysis for IMDB movie reviews

The training dataset can be loaded using read_csv(url(“https://yongjunzhang/files/css/imdb_sample.csv”)).

It is a random sample of IMDB movie review data. It has 1500 negative reviews and 1500 positive reviews.

You are required to compute sentiment scores using one of lexicon-based methods and one of conventional ML methods.

You need to turn in your rmarkdown file and the associated pdf or html.

Some clues: You can use one of these r packages we mentioned in our lecture and just do mutate to create a sentiment score. For ML method, you can do caret style training or use quanteda package.

You have two weeks to complete this challenge.

Note that you should also work on your final research project as well.