Contextual polarity and influence mining in online social networks

Abstract Crowdsourcing is an emerging tool for collaboration and innovation platforms. Recently, crowdsourcing platforms have become a vital firms to generate new ideas, especially large such as Dell, Microsoft, Starbucks, provides with multiple advantages, notably, rapid solutions, cost savings, variety of novel ideas that represent the diversity inherent within crowd. The literature on limited empirical evidence advantage businesses strategy. In this study, Starbucks’ platform, Ideas examined, three objectives: first, determine participants’ perception company by participants when generating platform. second objective map users into community structure identify those more likely produce ideas; most promising are grouped communities best ideas. third study relationship between users’ ideas’ sentiment scores frequency discussions among users. results indicate emotion can be used visualize social interaction narrative over time. They also suggest fast greedy algorithm one suited modularity agreeable 0.53 8 significant using edge weights. For disagreeable 0.47 without There statistically quadratic sentiments number conversations