Software-engineering-group-journals

Many companies store a lot of massive data and typically, the data itself contains a lot of non-trivial but useful information. Data mining techniques can be used to discover this information which can help the companies for decision-making. However, in real life applications, data is massive and is stored over distributed sites. One of my major research topics is to protect privacy over this kind of data. Protecting privacy over massive data stored in distributed sites is very important in our daily lives because there can be serious consequences if the data owner releases data without considering the protection of sensitive information. Some real events that demonstrate the problem are the following. The first event is that in 2002, Sweeney identified the insufficient protection of a medical dataset. In a real medical data, about 87% of individuals can be uniquely identified by matching certain attributes with a publicly available external table such as a voter registration list by a simple mapping operation. The second event is in 2006; AOL did not take sufficient precautions and encountered some undesirable consequences. A dataset including search logs was published in 2006. Later, AOL realized that a single 62 year old woman living in Lilburn (in Geogia) can be identified from the search logs by New York Times reporters using her several individual-specific queries (e.g., finding webpages with her last name and finding landscapers in Lilburn). The search logs were withdrawn and two employees responsible for releasing the search logs were fired . The third event comes from Netflix. Netflix is a popular online movie rental service with a recommender system, called Cinematic that recommends movies to its customers based on their predicted movie preferences. Netfix released its data to the public on October 2, 2006 for a challenging competition called Netflix Prize with the aim of improving the prediction accuracy of the recommender system. However Narayanan, found that 96% of the subscribers could be uniquely identified by limited knowledge of at most 8 movie ratings with their corresponding rating dates. The fourth event is that the research project conducted by Beinat shows that 24% of the mobile clients using location-based services (LBS) have serious privacy concerns about disclosing their locations together with their personal information. (Big Data Privacy, Raymond Chi-Wing Wong)

• Share this page
• 
• 
• 
• 
• 
• 

Last date updated on September, 2024