NCapture Twitter Word Frequencies
Access Jupyter Notebook on Google Colab here: https://colab.research.google.com/drive/1a-U8Bx1KvPrLaLNkM3eMJyJQ7183IchE?usp=sharing
Problem: NVivo sets all NCapture record columns from Twitter as Classifying and only the Tweet itself as Codable. For researchers wanting word frequencies and wordclouds from a different field, such as the hashtags, the optimal method is to (1) import into NVivo, (2) export as Excel, (3) re-import to NVivo using the Survey import tool, where other fields can be set as Open and therefore Codable.
Solution: This Jupyter Notebook tool parses NCapture files and generates a wordcloud and structured frequency table, including word lemmas, parts of speech tags, and named entities.
Non-Default Libraries Used
- spaCy – https://spacy.io/
- Wordcloud – https://github.com/amueller/word_cloud