Ethical AI and Librarianship
A Resource Guide
Computer Vision for the Humanities: An Introduction to Deep Learning for Image Classification (Part 2)
|
Field
|
Description | |
|---|---|---|
| Title | Computer Vision for the Humanities: An Introduction to Deep Learning for Image Classification (Part 2) | |
| Type | Courses & Tutorials | |
| Creator | Daniel van Strien, Kaspar Beelen, Melvin Wevers, Thomas Smits, Katherine McDonough | |
| Link | https://programminghistorian.org/en/lessons/computer-vision-deep-learning-pt2 | |
| Creation Date | 08/17/2022 | |
| Last Updated Date | 07/11/2024 | |
| Summary | This resource is Part 2 of a two-part online lesson introducing deep learning–based computer vision methods for humanities research. The lessons use a data set related to visual content extracted from over 16 million digitized historic newspaper pages from the Library of Congress’s Chronicling America collection. While Part 1 introduces how to build a basic image classification model and understand a deep learning pipeline, Part 2 builds on Part 1 by covering more steps in a deep learning pipeline, challenges related to training data, and importance of selecting appropriate evaluation metrics. Part 2 concludes with a reflection on the categories of classification, emphasizing how classification scheme inevitably abstracts away visual nuance. The authors encourage researchers to consider whether a chosen level of abstraction in the labeling categories provides meaningful insights about visual collection. The lesson ends with recommended readings, including resources on ethical issues in applying machine learning to library and cultural heritage contexts, such as “Responsible Operations: Data Science, Machine Learning, and AI in Libraries (2019), OCLC Research”. | |
| Topic | AI and Archive. Digital Collection. | |
| Source and Link | Programming Historian. https://programminghistorian.org/ | |
| Access | Open | |
| Accessibility | Open | |
| Audience | Information professionals. | |
| Platform or Format | Web | |
| Length | -- | |
| Geography | GBR/USA/Europe | |
| Language | ENG/FRE/SPA/PRT | |
| Description Date | 06/17/2025 |