Grant agreement: Academy of Finland no. 345610
Start date: 01-01-2022
Duration: 24 months
WP 3.4: Report on Livestream data collector
Date of reporting: 18-04-2023
Report author: Tanja Välisalo (JYU)
Contributors: Jari Lindroos (JYU), Raine Koskimaa (JYU), Jaakko Peltonen (TUNI), Tanja Välisalo (JYU)
Standalone collector with mockup GUI and CLI functionality will be published in https://github.com/pwcd/twitcher. Currently, a Streamlit-based version for running the collector on the web, saving the collected data to a Hugging Face data repository is running at https://pwcd-st-twitcher-home-t7f36f.streamlit.app/
The proliferation of streamed audio-visual content with textual communication features has made a significant change to the media landscape. Current research into livestream chat has mainly typically used qualitative methods and limited samples. There is a need to study large masses of online streams quantitatively.
This deliverable is a data collection tool that enables collecting large amounts of chat data from the livestream service Twitch. The tool is currently shared via GitHub but a visual user interface is under development. The user is responsible for the permissions and ethical choices related to collecting the data.