Social scientists may find that working with semi-unstructured data, such as natural language text, is an essential but time consuming and difficult part of a research project. However, there are some simple computational techniques that researchers can learn that can improve how they work with text data. These techniques can help researchers speed up and simplify their text analysis as well as make their research methods more transparently documented and reproducible.
This free series, organised by the UK Data Service, introduces core text-mining concepts and demonstrates some basic and advanced methods that can be customised to the needs of individual research projects.
Importing text data and basic preparations – 02 September 2020 - recording
Basic Natural Language Processing – 16 September 2020 - recording
Training a classifier for sentiment analysis – 23 September 2020 - recording
Extracting named entities and creating a social network – 30 September 2020 - recording
Each code demo works through coding examples line-by-line, explaining the logical and programmatic aspects of the methods being demonstrated. During each session you will be able to run the code yourself in real time without any installation on your machine. Each demonstration uses the popular Python programming language and lasts for roughly 30 minutes, followed by up to 30 minutes for Q&A and other interaction.