3. Topic Modeling and Theme Extraction

Activity 3.01: Performing Topic Modeling on a Set of Documents with Unknown Topics

Solution:

1. For this activity, we are going to use 1,000 movie review files. Navigate to the following link (or to your local directory where you have downloaded the GitHub files) to obtain the text data files that contain movie review comments: https://packt.live/3gISDZL. It is definitely better to download the GitHub repository rather than download 1,000 files by hand.
2. Navigate to the S3 dashboard at https://s3.console.aws.amazon.com/s3/home.
3. Click the bucket that you created earlier (in my case, it is "aws-ml-input-for-topic-modeling-20200301"):

   Figure 3.50: The S3 home screen for 'input-for-topic-modeling'

4. Click Create folder:

   Figure 3.51: Clicking on Create folder

5. Type movie_review_files and click Save:

   Figure 3.52: Clicking Save

   Note

   For this step, you may either follow along with the exercise and type in the code in...