Role of privacy attacks and utility metrics in crowdsourcing for urban data analysis

In current era, excessive usage of mobile devices and internet people often participate in the surveys, questionnaires, usability tests, performance measures and quantitative reviews. This process of outsourcing the data collection from the crowd is called mobile crowdsourcing. It involves large group of participating people and allows the researcher or analyst to gather data in real time at relatively lower cost when compared to the traditional methods of data collection. Mobile crowdsourcing has applications in idea generation, urban planning and urban mobility, public participation in problem solving and decision making, collective intelligence, crowd wisdom and human computation. There is a threat to individual's sensitive or personal information when the data is shared. Privacy preservation is a major concern in mobile crowdsourcing as enormous amount of data is being collected from the crowd and used for analytics, forecasting and decision making by extracting useful information. These data contain private or sensitive information related to individual/person who owns it. If the data is used in its original form, it may lead to privacy disclosure as it contains person-specific data. Hence, it is the duty of data curator to anonymize the data, before it is published for public use. The original data should be anonymized in such a way that, it should be very challenging for intruder to obtain sensitive information by means of any privacy attack model.