Navigation auf uzh.ch
Navigation auf uzh.ch
In cooperation with the Stop Hate Speech project led by alliance F, the project team has developed an automatic classification system for hate speech and toxic language. With the help of volunteers they optimized the algorithm to detect hate speech more easily.
All comments had been collected from online fora of Swiss news outlets. Some of them feature hate speech or toxic language, others do not. The assessment of hate speech and toxic language use is very subjective. In order to optimize the system, volunteers were sought to help rate the comments.
Volunteers were asked to read comments on online articles and determine whether or not they contain hate speech and toxic language. The data analysis was done on our Citizen Science Project Builder platform.
The data analysis is now completed. The goal of 1500 annotations (3 x 500 comments) was reached. The project team is now analyzing the data. Thank you to everyone who participated!
|
|
The Digital Democracy Lab researches the political implications of digital technology using computational social science methods. It focuses on topics such as political communication and public opinion, e-government and public administration, AI and governance, civic tech and political participation, the regulation of tech platforms, and state repression and surveillance. |
|
|
The Immigration Policy Lab (IPL) at ETH Zurich and Stanford University investigates the effects of immigration and integration policies. We aim to provide empirical answers to questions about what does – and does not – work in the realm of migration and integration policy, and to generate innovative solutions based on rigorous evidence. Using cutting-edge analytical tools, we provide policy-makers and practitioners insights that inspire policies and that benefits immigrants and host communities alike. |
|
|
"Stop Hate Speech" is a project initiated by alliance F, the umbrella organization of Swiss women's associations. In collaboration with the Digital Democracy Lab, the Immigration Policy Lab, and further partners it aims to document the prevalence of hate speech and test the effectiveness of various counterspeech strategies in Switzerland. To this end, the project has developed a hate speech detection algorithm and designs randomized field experiments to measure the effects of counterspeech. |
