HICSS '10 Proceedings of the 2010 43rd Hawaii International Conference on System Sciences
Investigating multi-label classification for human values
Proceedings of the 73rd ASIS&T Annual Meeting on Navigating Streams in an Information Ecosystem - Volume 47
Comparing values and sentiment using Mechanical Turk
Proceedings of the 2011 iConference
Comparing values and sentiment using Mechanical Turk
Proceedings of the 2011 iConference
Proceedings of the 2012 iConference
Hi-index | 0.00 |
An annotator's classification of a text not only tells us something about the intent of the text's author, it also tells us something about the annotator's standpoint. To understand authorial intent, we can consider all of these diverse standpoints, as well as the extent to which the annotators' standpoints affect their perceptions of authorial intent. To model human behavior, it is important to model humans' unique standpoints. Human values play an especially important role in determining human behavior and how people perceive the world around them, so any effort to model human behavior and perception can benefit from an effort to understand and model human values. Instead of training humans to obscure their standpoints and act like computers, we should teach computers to have standpoints of their own.