Reading

Machine Learning, Explained

Sara Brown

April 21, 2021

This readable, non-technical article, digitally penned by writer Sara Brown for the Ideas Made to Matter website of the MIT Sloan School of Management, offers a comprehensive view of machine learning (ML), including a discussion of the three main categories of ML: supervised, unsupervised, and reinforcement.  Brown also discusses some of the common ML technologies and algorithms, including artificial neural networks and deep learning.  Although written from a business perspective and including many examples of business applications, the article also explains many ML approaches that are directly relevant to the digital humanities, including natural language processing and image analysis.  Brown also emphasizes the importance of data in machine learning.

READ: Machine Learning, Explained

Machine Learning. An Introduction

Gavin Edwards

November 18, 2018

This longer and more detailed introduction to machine learning discusses the topic in an intuitive, relatively non-technical manner.  It also presents background theory and clarifies terminology.  Concepts that are important in the context of the digital humanities are also examined, including classification, clustering, regression, anomaly detection, and dimensionality reduction.

READ: Machine Learning. An Introduction

Which Machine Learning Algorithm Should I Use?

Hui Li

December 9, 2020

The SAS Data Science Blog

This blog site lists the main techniques in machine learning and discusses the different categories of machine learning algorithms.  The site provides a resource for data scientists and others in selecting appropriate machine learning algorithms to solve specific problems.  It features a “Machine Learning Algorithms Cheat Sheet” graphic to aid in the selection process.

VIEW: Which Machine Learning Algorithm Should I Use?

Parameters and Hyperparameters in Machine Learning and Deep Learning

Kizito Nyuytiymbiy

Dec 30, 2020

This article explains the difference between parameters and hyperparameters and the role of both in machine learning.

Read: Parameters and Hyperparameters in Machine Learning and Deep Learning

Python and R Code

The regression example presented in this section is based on the Python code: Regression_Example.py.  Because the data are generated randomly, the exact coefficient values may differ slightly.  A Jupyter Notebook (Regression_Example.ipynb) is available for this code.   The R code for the interactive visualization (GenderedPerspectives_Visualization.R) is available.  It uses the data file genderedPerspectives_data4r.csv.  A Jupyter Notebook for this visualization (GenderedPerspectives_Visualization.ipynb) is also available.