ABSTRACT
Massive network resources not only can be refined and developed into a useful classification system tools, in the actual implementation, data is ubiquitous and can be expanded of use. “Crowd labeling algorithm,” as a typical example, making big data active, is to give data vigor. “Crowd labeling algorithm” is to collect data through online crowd behavior, extract and mark product-oriented label, via the crowd label clustering, and analyzes the behavior of crowds and trends of interest (Yin 2012, Song & Zhang 2011).
