Once students become acquainted with meter, we turn to basic concepts of machine learning, again beginning with a glossary approach, reviewing terms like machine learning, artificial intelligence, feature, and target concept.
The stress markers are placed above the lines, and the foot separations (shown as uprights) are placed in them. The For Better for Verse online learning tool gives automated feedback on scansion. They visit the University of Virginia’s For Better for Verse web-based learning tool to explore the meter of an entire poem, and receive automated feedback on their scansion and insight into how meter might affect the poem’s meaning (Figure 1). Students then identify the stress patterns of words as a group, and ultimately perform scansion on lines of poetry to identify its meter. In “The Bardic Bot: Training AI to Recognize Poetic Meters,” we first introduce essential concepts of meter and line scanning using a glossary of important terms, including meter, scansion, syllabification, and stressed and unstressed syllables.
ONLINE SCANSION TOOL HOW TO
While building these skills can feel tedious and time-consuming, we believe that learning how to train machine learning models to identify the nuances of meter will engage students. Students must develop a collection of related skills: identifying syllables in words, understanding and labeling different units of meter (e.g., feet and terms for line length), and connecting these patterns with the poem’s meaning. Teaching meter to students is a complex process, especially when the goal is to develop competence in both writing and reading poetry. We designed a weeklong StoryQ curriculum module around iambic meter, the metrical mainstay popularized by Shakespeare and reflective of natural speech patterns in English. They acknowledge that “while the rhythm in most line encountered in a work of poetry appears mundanely repetitive on the surface, poetry, while mostly a constrained literary form, is prone to unexpected deviations of such standard patterns.” It is this continual setup and subversion of literary expectations that makes meter an ideal playspace for machine learning and provides an opportunity to teach AI fundamentals in the English classroom. In their 2016 paper for the International Conference on Computational Linguistics, Manex Agirrezabala, Iñaki Alegria, and Mans Hulden* apply Natural Language Processing (NLP) techniques to a selection of poetry in an attempt to identify its meter-the underlying rhythm expressed through stressed and unstressed syllables. Because public domain poetry texts are widely available and far shorter than novels, they make great candidates for introducing machine learning techniques in the ELA curriculum.
Our Narrative Modeling with StoryQ project aims to integrate AI into existing disciplinary studies such as English Language Arts (ELA) in order to prepare youth for the future.Īmong many literary genres that students encounter in high school, poetry presents a unique opportunity for integrating AI education. This means that many students simply write off a future in AI because they aren’t “math people” or don’t think they can learn how to code.
However, opportunities to study AI at the pre-college level, if available at all, are limited to computer science classes. From increasingly autonomous self-driving cars to climate change models, Artificial Intelligence (AI) has become a ubiquitous medium for understanding, explaining, and interacting with the world around us.
0 kommentar(er)
