{"id":137,"date":"2018-07-09T18:26:18","date_gmt":"2018-07-09T18:26:18","guid":{"rendered":"http:\/\/sarahjpurcell.sites.grinnell.edu\/sandbox_clone\/?page_id=137"},"modified":"2019-10-30T16:50:47","modified_gmt":"2019-10-30T16:50:47","slug":"topic-modeling","status":"publish","type":"page","link":"https:\/\/his100.sarahjpurcell.sites.grinnell.edu\/spring-2021\/tutorials\/textual-analysis\/topic-modeling\/","title":{"rendered":"Introduction to Topic Modeling"},"content":{"rendered":"

This tutorial was written by Katherine Walden<\/a>, Digital Liberal Arts Specialist at Grinnell College.<\/p>\n

This tutorial was reviewed by Sarah Purcell<\/a> (L.F. Parker Professor of History) and Gina Donovan<\/a> (Instructional Technologist) at Grinnell College, and edited by Papa Ampim-Darko, a student research assistant at Grinnell College.<\/p>\n

This tutorial is adapted from the Programming Historian\u2019s Topic Modeling and MALLET tutorial<\/a>.<\/p>\n

\"Creative<\/a>
\nIntroduction to Topic Modeling is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License<\/a>.<\/p>\n

———-<\/p>\n

As the authors in Exploring Big Historical Data <\/em>point out, \u201cKeywords have their limitations, in that they require us to know what to search for\u201d (113). As an alternative, topic modeling builds a list of significant terms by calculating what terms, phrases, clusters, etc. are significant or occur frequently within a text.<\/p>\n

In this tutorial, we will introduce topic modeling using digital tools that represent the types of technical interfaces historical scholars use in their research. Across these digital tools, we will be using the process Underwood describes for topic modeling:<\/p>\n

\u201cYou assign words to topics randomly and then just keep improving the model, to make your guess more internally consistent, until the model reaches an equilibrium that is as consistent as the collection allows\u201d (Underwood, quoted on page 119)
\nExploring Big Historical Data<\/em>: Chapter 4<\/p><\/blockquote>\n

\n

GUI Topic Modeling Tool (GTMT)<\/h4>\n

As we\u2019ll explore later in this tutorial, the more robust topic modeling tools can sometimes have a steep technical learning curve. The Topic Modeling Tool<\/a> in a Google Code repository provides a graphical user interface for preliminary topic modeling. The project was created in 2011 through a Institute of Museum and Library Services grant and was developed by Yale University, the University of Michigan, and the University of California, Irvine. The GUI Topic Modeling Tool is built on the MALLET toolkit<\/a>.<\/p>\n

\n
Data<\/h5>\n

Navigate to https:\/\/sarahjpurcell.sites.grinnell.edu\/digital_methods\/files\/Equiano_Text.txt <\/a>in a web browser and save the \u201cFiles\u201d compressed zip folder to your Desktop.<\/p>\n

Extract the folders contents.<\/p>\n

 <\/p>\n

\n
Launching GTMT<\/h5>\n

GTMT is already installed on the computers in our classroom.<\/p>\n

To use it on your own computer, go to the GitHub page<\/a>, download the Mac or Windows version, and follow the installation instructions.<\/p>\n

\"\"<\/a><\/p>\n

Navigate to C:\\Program Files (x86)\\TopicModelingTool\\app<\/strong> in File Explorer, and double click on the TopicModelingTool-jfx<\/strong> file to launch the program.<\/p>\n

\n
Importing and Analyzing Data<\/h5>\n

\"\"<\/a><\/p>\n

1-Click the Input Dir\u2026<\/strong> icon and select the folder where the Equiano text file is located.<\/p>\n

2-Click the Output Dir\u2026<\/strong> icon and create a folder<\/strong> on the Desktop named Topic Modeling<\/strong>. Select this folder as your Output Directory.<\/p>\n

3-Click the Learn Topics<\/strong> icon to topic model the source text.<\/p>\n

\"\"<\/a><\/p>\n

4-What you see happening in the Console is the back-end work the MALLET tool does to generate the topic models.<\/p>\n

5-Once modeling is complete, the locations for output files is the folder you selected as Output Dir\u2026 The full file path for the output files is listed under **Generating Output**<\/strong> in the Console.<\/p>\n

\n
Exploring Results<\/h5>\n

\"\"<\/a><\/p>\n

6-Open File Explorer<\/strong> and navigate to the folder you selected as the Output Directory<\/strong>. In that folder, you have the results of your topic modeling as CSV and HTML files<\/strong>.<\/p>\n

7-Click on the output_html folder<\/strong>.<\/p>\n

8-The Docs folder<\/strong> contains an HTML file with topic modeling for each individual document contained in the Input Directory folder<\/strong>. All_topics<\/strong> is an HTML file that provides the list of topics. The Topics folder<\/strong> contains an HTML file that lists how many times a specific topic appears in each document.<\/p>\n

\"\"<\/a><\/p>\n

9-Open the all_topics HTML file<\/strong>.<\/p>\n

10-Based on your existing knowledge of the source text, how accurate or expected\/believable are these topics? What terms or topics surprise you? What terms or topics are absent that you think might be significant?<\/p>\n

11-Open the output_html folder<\/strong> and open the HTML file for a specific topic.<\/p>\n

12-Based on your knowledge of the source text, are you surprised by the distribution of this topic or where it appears in the text? What questions do you have about the results for this topic?<\/p>\n

13-Open the output_csv folder<\/strong>. The topics-words CSV<\/strong> includes the list of topics generated from the GTMT.<\/p>\n

14-The topics-in-docs CSV file<\/strong> is organized by document and lists the topics that appear in each document. The docs-in-topics CSV file<\/strong> is organized by topic and lists the documents that include each topic.<\/p>\n

15-Open both CSV files<\/strong>. How does the organization of information compare across the two table structures? How is the presentation of information in the CSV files different than your experience with the HTML files? What questions do you have about the information contained in the CSV files?<\/p>\n

16-How useful did you find the GTMT for enriching your understanding of the text?<\/p>\n

\n
Additional Exploration<\/h5>\n