首页|Data from Swiss Federal Institute of Technology Zurich (ETH) Advance Knowledge i n Machine Learning (Where you go is who you are: a study on machine learning bas ed semantic privacy attacks)
Data from Swiss Federal Institute of Technology Zurich (ETH) Advance Knowledge i n Machine Learning (Where you go is who you are: a study on machine learning bas ed semantic privacy attacks)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Current study results on artificial in telligence have been published. According to news originating from the Swiss Fed eral Institute of Technology Zurich (ETH) by NewsRx editors, the research stated , "Concerns about data privacy are omnipresent, given the increasing usage of di gital applications and their underlying business model that includes selling use r data." Funders for this research include Swiss Federal Institute of Technology Zurich. Our news editors obtained a quote from the research from Swiss Federal Institute of Technology Zurich (ETH): "Location data is particularly sensitive since they allow us to infer activity patterns and interests of users, e.g., by categorizi ng visited locations based on nearby points of interest (POI). On top of that, m achine learning methods provide new powerful tools to interpret big data. In lig ht of these considerations, we raise the following question: What is the actual risk that realistic, machine learning based privacy attacks can obtain meaningfu l semantic information from raw location data, subject to inaccuracies in the da ta? In response, we present a systematic analysis of two attack scenarios, namel y location categorization and user profiling. Experiments on the Foursquare data set and tracking data demonstrate the potential for abuse of high-quality spatia l information, leading to a significant privacy loss even with location inaccura cy of up to 200 m. With location obfuscation of more than 1 km, spatial informat ion hardly adds any value, but a high privacy risk solely from temporal informat ion remains. The availability of public context data such as POIs plays a key ro le in inference based on spatial information."
Swiss Federal Institute of Technology Zu rich (ETH)CyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
