This toolkit contains tools to extract conversational features and analyze social phenomena in conversations. Several large conversational datasets are included together with scripts exemplifying the use of the toolkit on these datasets.
The toolkit currently implements features for:
-
Linguistic coordination, a measure of relative power between individuals or groups based on their use of function words (see the Echoes of Power paper)
-
Question typology, a method for extracting surface motifs that recur in questions, and for grouping them according to their latent rhetorical role (see the Asking too much paper)
-
Coming soon: Politeness, currently available here: Politeness API
-
Coming soon: Basic message and turn features, currently available here Constructive conversations
The code for the toolkit can be found here.
Installation
- Install or use
Python 3. - Run
python3 setup.py installto install the package. - Run
python -m spacy download en
Use
Use import convokit to import it into your project.
Detailed installation and usage examples are also provided on the specific pages dedicated to each function of this toolkit.
Documentation is hosted here.
The documentation is built with Sphinx (pip3 install sphinx). To build it yourself, navigate to doc/ and run make html.
Andrew Wang (azw7@cornell.edu) wrote the Coordination code and the respective example script, wrote the heper functions and designed the structure of the toolkit.
Ishaan Jhaveri (iaj8@cornell.edu) refactored the Question Typology code and wrote the respective example scripts.