Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Quantitative intertextuality : analyzing the markers of information reuse
Forstall C., Scheirer W., Springer Publishing Company, Incorporated, New York, NY, 2019. 189 pp. Type: Book (978-3-030234-13-3)
Date Reviewed: Nov 26 2019

Intertextuality is a term used to denote the idea that there are relationships between texts. This includes the conscious modeling of one text on another, allusions to phrases from other texts, and even plagiarism. The goal of this book is to introduce readers to the use of digital methods, such as machine learning, in the study of intertextuality. The authors use the phrase “quantitative intertextuality” for this discipline. The work builds on the Tesserae project (https://tesserae.caset.buffalo.edu/), which began as a study of the influence of Virgil’s Aeneid on Lucan’s Civil War (also know as the Pharsalia).

Forstall is a classics professor and Scheirer is a computer science professor. The result is a somewhat unusual approach to the presentation of ideas. Each chapter (apart from the first two) begins with an overview of the material, emphasizing a humanities approach to the topic. This is followed by a detailed case study, including a quantitative model and illustrative code written in R that implements the ideas.

Chapter 1 defines the notion of quantitative intertextuality in more detail, for example, the identification of measurable features of texts, the identification of patterns among the features, and the discovery of meaningful relationships between the texts based on these features. The chapter gives some examples of earlier projects along these lines. Chapter 1 thus announces the aims of the book.

Chapter 2 reviews previous work on intertextuality, again mentioning examples such as the structural references to Homer’s Odyssey in Joyce’s Ulysses and the Coen brothers’ film O Brother, Where Art Thou? The chapter concludes with a section introducing machine learning as a tool for intertextuality.

Chapters 3 through 7 form the heart of the book in that they are case studies of intertextuality, with code examples that implement the ideas. Chapter 3 is based on the use of words as the basic unit of intertextuality. The example here is allusions on Twitter to the HBO show Game of Thrones. In particular, the authors trace allusions to the phrase, “The Lannisters send their regards.” An important component of this example is the ranking of the potentially allusive phrases discovered by the system.

Chapter 4 considers semantic matching that traces reuse by meaning. The main tool used here is latent semantic indexing to capture intertextuality. The main example uses two novels from Wattpad, an Internet writing community.

Chapter 5 addresses sound matching. The context here is poetry and the influence of the English poet William Cowper on William Wordsworth. The main tool is the selection and tracing of character-level n-grams (er, it, ri, st, and so on).

Chapter 6 turns to the reuse of visual elements as a form of intertextuality. Here the major tool is the histogram of oriented gradients. The ideas are applied to Internet memes.

Chapter 7 discusses supervised machine learning, particularly the W-SVM classifier. The case study used here is self-plagiarism. This is not always reprehensible because the critic may be able to determine the degree to which the self-plagiarism is polishing rather than just copying.

Chapter 8 offers reflections on the material covered, while at the same time acknowledging that more powerful machine learning tools are available and should be used in the future.

Each chapter includes a comprehensive bibliography at the end. Chapters 3 through 7--the case study chapters that contain code--include student exercises. All of the code is available online, as is some of the data (not all is available because of copyright issues, though references for any such data are given). The code examples are lucidly explicated so that the reader gains a clear understanding of the functions of the code. Some footnotes give warnings about possible problems with the precise implementation details for various instances of R, a most welcome intervention by the authors.

As a work that bridges two traditionally distinct disciplines, the book is most welcome and can benefit both humanities scholars and so-called hard scientists.

Reviewer:  J. P. E. Hodgson Review #: CR146800 (2004-0069)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Learning (I.2.6 )
 
 
Computer Vision (I.5.4 ... )
 
 
Natural Language Processing (I.2.7 )
 
 
Artificial Intelligence (I.2 )
 
 
General (I.0 )
 
Would you recommend this review?
yes
no
Other reviews under "Learning": Date
Learning in parallel networks: simulating learning in a probabilistic system
Hinton G. (ed) BYTE 10(4): 265-273, 1985. Type: Article
Nov 1 1985
Macro-operators: a weak method for learning
Korf R. Artificial Intelligence 26(1): 35-77, 1985. Type: Article
Feb 1 1986
Inferring (mal) rules from pupils’ protocols
Sleeman D.  Progress in artificial intelligence (, Orsay, France,391985. Type: Proceedings
Dec 1 1985
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy