Text Analysis II: Exploring Voyant Tools with an AI-Generated Sample
— by Katherine Choi, Kayla Ng, Terry Chung
Last time we briefly introduced text analysis and its applications. In this post, we will explore text analysis using Voyant Tools, a web-based platform known for its analytical capabilities. We will be working with an AI-generated short novel, “A Whisper in the Wind”, as our sample text. Together, we will explore key themes, apply research methodologies, and guide you step-by-step through the process of using Voyant Tools to uncover insights effectively.
Introduction to Voyant Tools
Voyant Tools is a web-based text analysis platform that requires no initial setup. Users can either upload text data directly for analysis or set up a local server for more secure or large-scale projects, with clear instructions provided. This flexibility makes Voyant Tools a popular choice for both beginners and advanced users in text analysis.
Preparation for Analysis
1. Accessing Voyant Tools
- Visit https://voyant-tools.org/.
Figure 1: The Homepage of Voyant Tools
2. Creating a Corpus
Option 1: Copy and paste your text directly into the text box. This is a quick way to get started if you have a manageable amount of text.
Option 2: Paste URLs of web sources (one per line) into the text box. This allows you to analyze content from online sources directly.
Figure 2: Creating a corpus by pasting URLs into Voyant Tools
- Option 3: Upload text files from your local computer (Word, PDF, TXT, etc.). This is useful for analyzing larger documents or files you’ve already prepared.
Figure 3: Creating a corpus by uploading text file to Voyant Tools
For this demonstration, we will use Option 1 and paste our sample text into the text box to create our corpus.
Exploring the Default Interface
When you create or open a corpus in Voyant Tools, you will see its default interface, which includes five panels:
- Cirrus Panel: Visualizes frequently used words in a word cloud.
- Reader Panel: Displays the text of your corpus, allowing interaction with the other panels.
- Trends Panel: Shows frequency trends of selected words throughout the text.
- Summary Panel: Provides an overview of the main statistics of the corpus.
- Contexts Panel: Shows the context and collocation of specific terms.
These panels provide a comprehensive overview and various ways to analyze your text, but they are just the starting point. You can customize the panels and tools to fit your specific research needs.
Figure 4: The default interface of Voyant Tools
Research Question
Our analysis aims to address the following questions:
Q1: What are the most frequently occurring themes and terms in the novel?
Q2: How does the conflict develop and resolve throughout the story?
Techniques Used
Cirrus Panel
- Purpose: Visualize the most commonly occurring words in the text.
- Method: Creates a word cloud where word size reflects its frequency.
Figure 5: The word cloud for “A Whisper in the Wind” is displayed in the Cirrus Panel
Trends Panel
- Purpose: Examine how the frequency of specific words changes over the course of the text.
- Method: Displays a line graph that tracks the relative frequency of selected words across different parts of the corpus.
Figure 6: The trend graph for “A Whisper in the Wind” is displayed in the Trend Panel
Using the Cirrus Panel
- Adjusting Visualization: Adjust the number of words displayed to focus on the most significant terms.
- Filtering Words: Edit the stop word list to filter out unwanted words, refining the word cloud’s focus.
Figure 7: The Cirrus Panel settings
Using the Trends Panel
- Segmentation: Break the text into smaller or larger parts for better analysis in different parts.
- Selecting Keywords: Choose keywords found in the word cloud to analyze their frequency trends.
- Interpreting Trends: Use the trends to understand the development and emphasis of different themes and elements.
Figure 8: The Trends Panel settings
Findings
Q1: What are the most frequently occurring themes and terms in the novel?
Cirrus Panel Findings
- Key Words Identified: The word cloud highlights the most frequently used words in the novel. Prominent words include “eva”, “mansion”, “curse”, “clara”, “wind”, “chapter”, “secrets”, “hidden”, and “discovered”.
- Interpretation: The prominence of terms like “eva”, “mansion”, “curse”, and “clara” indicates that these are central to the storyline. The frequent appearance of words like “secrets”, “hidden”, and “discovered” suggests themes of mystery and discovery.
Figure 9: The word cloud from the Cirrus Panel
Q2: How does the conflict develop and resolve throughout the story?
Trends Panel Findings
Analysis of Key Words:
- Mansion: High frequency in the beginning, tapering off mid-story, and resurging at the end. The mansion is established as the central setting where the curse originates and ultimately resolves.
- Curse: Frequency rises, peaking during the middle sections, marking the intensification of the conflict.
- Eva: Mentioned consistently, with a slight increase toward the climax, underscoring her crucial role in confronting and resolving the curse.
- Wind: Moderate frequency, increasing during climactic moments. The wind symbolizes the chaos and intensity of the curse.
- Clara: Maintains a steady presence across the story, playing a supporting but significant role
Detailed Insights:
- Introduction and Rising Action (Bins 1-2):
- The mansion is frequently mentioned early on, setting the stage for the curse (conflict).
- The curse begins to rise in frequency, signaling its introduction and growing significance in the plot.
- Eva and Clara are both frequently mentioned, indicating their central roles in the early development of the conflict.
- Climax (Bins 3-4):
- The curse reaches its peak frequency, reflecting the height of the conflict. The focus shifts from the mansion to the direct impact of the curse on the characters.
- Eva becomes more prominent, emphasizing her role in confronting the curse.
- The wind also rises, symbolizing the intensifying and chaotic nature of the curse.
- Falling Action and Resolution (Bins 5-6):
- The curse frequency declines, indicating the resolution of the conflict.
- Eva remains central, reflecting her importance in overcoming the curse.
- The return of mansion mentions suggests a restoration of normalcy, while Clara continues her steady presence, supporting the resolution.
Figure 10: The refined trend graph displays the trends of “mansion,” “curse,” “Eva,” “wind,” and “Clara” throughout the story
Limitations of Voyant Tools
Voyant Tools, while powerful, has several limitations. Users may encounter stability issues, and the stopword list may not function as expected after updates. Interactions between panels can be tricky and require patience to master. The tool’s tokenization method might not always be perfectly accurate, and its web-based nature can lead to occasional connectivity problems. Despite these challenges, Voyant Tools remains a beginner-friendly platform that significantly lowers the barrier for those new to text analysis. Reporting issues on its GitHub page is encouraged to help improve the tool.
Conclusion
This post demonstrated the use of Voyant Tools for text analysis with an AI-generated short novel, A Whisper in the Wind, as a case study. Voyant Tools offers various functionalities for text analysis through panels like Cirrus and Trends, which help visualize word frequencies and trends. The demonstration aimed to illustrate how Voyant Tools can be used to identify and examine key themes and patterns in a text. While the tool offers a straightforward interface suitable for both beginners and more experienced users, it does have some limitations, including potential stability issues and occasional challenges with tokenization accuracy. Despite these limitations, Voyant Tools remains a valuable resource for text analysis and serves as an accessible entry point for more complex analytical tasks.
Extended Reading
Appendix: The sample AI-generated novel
A Whisper in the Wind
Chapter 1: The Arrival
Eva Morgan stepped off the train and into the sleepy village of Pinecliff. Tall pines framed the horizon, and the scent of earth and pine needles filled the crisp autumn air. She clutched a worn letter from her great-grandmother, Clara, inviting her to stay at the ancestral home she had never seen.
The villagers watched as Eva walked to the old mansion at the edge of town, a grand structure that had been both admired and feared for generations. The house, with its towering turrets and ivy-clad walls, seemed to whisper of secrets long buried.
Chapter 2: The Discovery
The inside of the mansion was a labyrinth of shadowed hallways and creaky floorboards. Eva explored with a mix of trepidation and curiosity. In the library, she found an old diary bound in cracked leather. It belonged to Clara.
Opening the diary, Eva read about the hidden room in the mansion, a place where the family’s greatest secrets were kept. Clara’s words hinted at a treasure, but also a curse that had plagued the Morgan family for centuries.
Chapter 3: The Hidden Room
Following Clara’s cryptic clues, Eva discovered a concealed door behind an aging tapestry. Inside, the room was untouched by time. Shelves lined with ancient books, artifacts from distant lands, and a grand desk holding a locked chest.
Eva found an ornate key hidden beneath a loose floorboard. Her hands trembled as she unlocked the chest. Inside lay a golden locket and a faded photograph of Clara with a man Eva didn’t recognize.
Chapter 4: The Curse
Eva returned to the diary, where Clara detailed her love for a man named Alistair, a romance shrouded in mystery. Their love had been the reason for the family curse, cast by a jealous rival who sought to destroy their happiness.
Eva felt a chill as she heard whispers in the wind, as if Clara’s spirit was speaking to her. Determined to break the curse, she sought out the village’s historian, an elderly man named Mr. Whitaker.
Chapter 5: The Revelation
Mr. Whitaker revealed that the curse could only be broken by reuniting the locket with a similar one worn by Alistair. According to legend, Alistair had been buried in an unmarked grave in the village cemetery.
Eva and Mr. Whitaker ventured to the cemetery at dusk. Guided by Clara’s whispers, they found the grave and uncovered Alistair’s locket. As Eva placed the two lockets together, a gust of wind swirled around them, lifting the curse.
Chapter 6: The New Beginning
The mansion seemed to brighten, as if the shadows held for generations were finally lifting. Eva felt a sense of peace as Clara’s voice whispered her gratitude. With the curse broken, she knew the mansion could become a place of joy once more.
Eva decided to stay in Pinecliff, turning the ancestral home into a haven for travelers and a library preserving the Morgan family’s rich history. The villagers welcomed her warmly, and the mansion became a vibrant part of the community.
The whispers in the wind no longer carried secrets and sorrows but tales of love, courage, and the promise of new beginnings.
[The sample text, “A Whisper in the Wind,” was generated by AI and is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0). You may share this work with proper attribution, but it may not be used for commercial purposes or adapted in any way. For more details, please visit the Creative Commons website (https://creativecommons.org/licenses/by-nc-nd/4.0/).]
