The enormous amount of online users, of diverse backgrounds, act as powerful resources of which Mobile Social Networks (MSNs) can utilize for crowdsourcing. Exploiting these online users as crowd workers is promising yet nontrivial. To efficiently leverage human intelligence or crowd wisdom, we need to address the following issues: 1) how to motivate users to participate, and 2) how to discourage malicious behaviors such as copying answers or making guesses. Furthermore, as low-quality answers may acutely degrade the accuracy of synthetic results, the last issue is 3) how to weed these out. In this thesis, we present MacroWiz, a simple yet effective platform to manage crowd wisdom on MSNs. Given a task, MacroWiz motivates online users to contribute their knowledge or opinions, an...[ Read more ]

The enormous amount of online users, of diverse backgrounds, act as powerful resources of which Mobile Social Networks (MSNs) can utilize for crowdsourcing. Exploiting these online users as crowd workers is promising yet nontrivial. To efficiently leverage human intelligence or crowd wisdom, we need to address the following issues: 1) how to motivate users to participate, and 2) how to discourage malicious behaviors such as copying answers or making guesses. Furthermore, as low-quality answers may acutely degrade the accuracy of synthetic results, the last issue is 3) how to weed these out. In this thesis, we present MacroWiz, a simple yet effective platform to manage crowd wisdom on MSNs. Given a task, MacroWiz motivates online users to contribute their knowledge or opinions, and assists the task holder in collecting answers, selecting the reliable ones, and making ultimate decisions. The platform consists of two functional units: online wisdom collection and offline answer selection. The former estimates and gathers the minimum number of answers required to satisfy the task requirement, while the latter analyzes the accuracy, effectiveness, and cost of each answer. Based on their suggestions those with high accuracy and low cost are selected by solving a double target optimization problem. We validate the effectiveness of our platform using MovieLens Data sets which contain over one million anonymous ratings of movies. Our result shows that this platform significantly reduces the latency in making decisions and provides high-quality answers at low cost.