Introduction to Digital Humanities

Text mining centers on identifying patterns and trends in unstructured texts. This often involves using a program or software to “read” text files and provide data about them, including data on word frequencies, common word patterns, tone indicators, and more. It is sometimes referred to as a "distant reading" method, in which you take a step back to identify patterns in language across a large group of texts. 

Many research questions and methods fall within the scope of text and data mining, including: 

• Identifying word frequencies
• Concordance (what passages mention specific key terms)
• Keyness (how often key terms appear in certain texts when compared to others) 
• Topic modelling (grouping key terms together to identify common themes and topics) 
• Named entity recognition (identifying names of people, places, things across texts)
• Sentiment analysis (identifying positive or negative tone)

For more advanced text mining techniques, such as sentiment analysis (identifying the tone of a text or texts) or named entity recognition (identifying people, places, and names in a text or texts), researchers often have to code their own text mining environments. R and Python are two commonly used programming software for text mining. Further resources for using programming software for text mining are linked below.