Machine Learning

Resources

• What 300-Dimensional Fridges Can Tell Us about Language

  EN

  In this lecture from the Austrian Centre for Digital Humanities and Cultural Heritage (ACDH-CH), Dirk Hovy gives an introduction to the method called embeddings, and showcases several applications of it. Hovy shows how they capture regional variation at an intra- and interlingual level, how they distinguish varieties and linguistic resources, and how they allow for the assessment of changing societal norms and associations.

• Understanding and Creating Word Embeddings

  EN

  Word embeddings allow you to analyze the usage of different terms in a corpus of texts by capturing information about their contextual usage. Through a primarily theoretical lens, this lesson will teach you how to prepare a corpus and train a word embedding model. You will explore how word vectors work, how to interpret them, and how to answer humanities research questions using them.

• The CLS INFRA Survey of Methods in Computational Literary Studies

  EN

  This resource from the CLS INFRA project offers an introduction to several research areas and issues that are prominent withinComputational Literary Studies (CLS), including authorship attribution, literary history, literary genre, gender in literature, and canonicity/prestige, as well as to several key methodological concerns that are of importance when performing research in CLS.

• Interrogating a National Narrative with GPT-2

  EN

  In this lesson, you will learn how to apply a Generative Pre-trained Transformer language model to a large-scale corpus so that you can locate broad themes and trends within written text.

• Facial Recognition in Historical Photographs with Artificial Intelligence in Python

  EN

  In this lesson, you'll learn computer vision and machine learning principles for object recognition, and how to apply these principles using Python to recognize and classify smiling faces in historical photographs.

• Creating Deep Convolutional Neural Networks for Image Classification

  EN

  This lesson provides a beginner-friendly introduction to convolutional neural networks (CNNs) for image classification. The tutorial provides a conceptual understanding of how neural networks work by using Google's Teachable Machine to train a model on paintings from the ArtUK database. This lesson also demonstrates how to use Javascript to embed the model in a live website.

• Computer Vision for the Humanities: An Introduction to Deep Learning for Image Classification (Part 2)

  EN

  This is the second of a two-part lesson introducing deep learning based computer vision methods for humanities research. This lesson digs deeper into the details of training a deep learning based computer vision model. It covers some challenges one may face due to the training data used and the importance of choosing an appropriate metric for your model. It presents some methods for evaluating the performance of a model.

• Computer Vision for the Humanities: An Introduction to Deep Learning for Image Classification (Part 1)

  EN

  This is the first of a two-part lesson introducing deep learning based computer vision methods for humanities research. Using a dataset of historical newspaper advertisements and the fastai Python library, the lesson walks through the pipeline of training a computer vision model to perform image classification.

• Clustering and Visualising Documents Using Word Embeddings

  EN

  This lesson uses word embeddings and clustering algorithms in Python to identify groups of similar documents in a corpus of approximately 9,000 academic abstracts. It will teach you the basics of dimensionality reduction for extracting structure from a large corpus and how to evaluate your results.

• Word Embeddings

  EN

  Natural language processing is one of the most powerful concepts in modern linguistics and computer science, bridging the understanding of language from human to machine, and in turn programming machines so they can perform complex linguistic tasks on their own. This short video introduces learners to the key concepts of word embeddings and how they can be used in digital humanities projects.

• Designing the Backbone of the RHDT using AI and Network Science

  EN

  This resource discusses how the ARTEMIS project combines artificial intelligence, network science, knowledge graphs, and semantic ontologies to transform unstructured cultural heritage data into structured, machine-readable knowledge supporting Reactive Heritage Digital Twins (RHDTs).

• Automatic Text Recognition (ATR) - Where and How to Get Images

  EN

  This tutorial explores where and how to find, create, and collect images of textual material, a crucial initial step in any process using Automatic Text Recognition (ATR).

• Automatic Text Recognition (ATR) - Getting Started

  EN

  Kick off your journey into Automatic Text Recognition (ATR) with our introductory tutorial video. This is the first video of a tutorial series dedicated to extracting full text from scanned images.

• Automatic Text Recognition (ATR) - Text Recognition and Post-ATR Correction

  EN

  Dive into the fine-tuning of Automatic Text Recognition outputs!

• Automatic Text Recognition (ATR)- Pre-Processing and Image Optimisation

  EN

  Get ready to improve the quality of your scanned images before moving to the processing phase of your ATR project.

• Automatic Text Recognition (ATR) - Layout Analysis

  EN

  Discover the subtleties of region and line segmentation and learn about the purpose of layout analysis for Automatic Text Recognition!

• Automatic Text Recognition (ATR) - End Formats and Reusability

  EN

  Increase the visibility of your ATR output while fostering Open Science.