What is DSI?
DSI (Divergent Semantic Integration) is an algorithm that estimates the creativity of English narratives or stories by computing the semantic distance between all words in the text (read the paper). While it has been validated using the Short Story task, DSI can also be applied to any text-based task to assess creativity.
How do I use DSI?
To obtain DSI scores for your text data, upload a cleaned (no blank responses) CSV file that has a column named 'response' containing the texts you want to analyze. The algorithm is case-sensitive for the response column name, but it does not matter if other columns are present in the input CSV.
The algorithm will process your input CSV and return it with a new column named 'DSI', which contains the calculated DSI scores for each text entry. Please note that computing DSI scores can be time-consuming, especially for longer texts or larger datasets.
While DSI is designed for English texts, it should perform reasonably well if the input texts are translated to English using Google Translate. However, it is not recommended to compute DSI scores directly in other languages.
