Crowd Worker Selection with Wide Learning and Narrow Evaluation

Abstract

Most people can easily find any place with enough portable devices and big data. Location information must already be known to someone, verified, and provided by a trusted provider. Therefore, Location Service Providers (LSP) may offer their clients biased information to use all of this information correctly and appropriately. But can clients are sure which LSP's approach is right for them? Therefore, it is very difficult to fit individuality into these tasks. We are attempting to solve this problem using collective intelligence to balance of information that is lacking in the Big Data industry. In our focus, Crowd Based System utilizes crowd wisdom to provide a variety of analytics. So using Worker Search Model (WSM) using learning techniques and Response Limit Model (RLM), which is a data selection set, we propose a strategy to optimize various interpretations to users. In addition, we challenge to find suitable locations by driving simulation. Simulation results show that our proposed system is about 1.5 times more likely to find a suitable worker compared to a simple conditional change approach.