Labs Overview

These versions of the Labs contain additional annotations to the Rmarkdown files. The annotations emphasize some key aspects of text analaysis, processing, coding, and document design and are indended to help support your progress.

Warning

What is in this documentation is not identical to what is in the Rmarkdown. For one, it omits the tasks, which you are asked to complete and submit. These are not included as those change from course to course.

Also, these supplementary materials were prepared using simple markdown (rather than Rmarkdown). Thus, their details and layout are not identical.

Labs

• Set 01
  • Lab 1: Texts, algorithms, and black-boxes
    • Load the cmu.textstat package
    • Prep the data and calculate sentiment
    • Transforming the data
    • Transformed vs. non-transformed data
    • What questions does this example raise for you?
  • Lab 2: The basics
    • A simple processing pipeline
    • What counts as a token?
  • Lab 3: Processing piplines
    • Tokenizing with quanteda
    • Tokenizing options
    • Creating a corpus composition table
• Set 02
  • Lab 4: Distributions
    • Prepare a corpus
    • Tokenize the corpus
    • Multi-word Expressions
    • Create a document-feature matrix
    • Token distributions
    • Distributions of the and of
    • Dispersion
    • Zipf’s Law
  • Lab 5: Collocations
    • Load the needed packages
    • Prepare the data
    • Collocates by mutual information (MI)
    • Create a tbl_graph object for plotting
    • Plot network
    • Reading in local files
  • Lab 6: Keyness
    • What is keyness?
    • Prepare a corpus
    • A corpus composition table
    • Keyness in quanteda
    • Creating sub-corpora
    • Effect size
    • Key key words
• Set 03
  • Lab 7: Part-of-speech tagging and dependency parsing
    • What does udpipe do?
    • Parse a sample sentence online
    • Basic parse structure
    • Visualize the dependency
    • Load the needed packages
    • Parsing
    • Preparing a corpus
    • Annotate a corpus
    • Process with quanteda
  • Lab 8: Logistic Regression
    • Load the needed packages
    • Case 1
    • Case 2
  • Lab 9: Multi-Dimensional Analysis
    • Case 1: Biber Tagger
    • Case 2: DocuScope
• Set 4
  • Lab 10: Cluster Analysis
    • Hierarchical Agglomerative Clustering
    • K-means