Expertise and Dynamics within Crowdsourced Musical Knowledge Curation: A Case Study of the Genius Platform

Many platforms collect crowdsourced information primarily from volunteers. As\nthis type of knowledge curation has become widespread, contribution formats\nvary substantially and are driven by diverse processes across differing\nplatforms. Thus, models for one platform are not necessarily applicable to\nothers. Here, we study the temporal dynamics of Genius, a platform primarily\ndesigned for user-contributed annotations of song lyrics. A unique aspect of\nGenius is that the annotations are extremely local -- an annotated lyric may\njust be a few lines of a song -- but also highly related, e.g., by song, album,\nartist, or genre. We analyze several dynamical processes associated with lyric\nannotations and their edits, which differ substantially from models for other\nplatforms. For example, expertise on song annotations follows a "U shape" where\nexperts are both early and late contributors with non-experts contributing\nintermediately; we develop a user utility model that captures such behavior. We\nalso find several contribution traits appearing early in a user's lifespan of\ncontributions that distinguish (eventual) experts from non-experts. Combining\nour findings, we develop a model for early prediction of user expertise.\n