|Dr Ahmed K. Elmagarmid is a computer scientist, academic, and executive. He is currently the Executive Director of Qatar Computing Research Institute in Doha, Qatar. Until 2010, he was a full Professor of Computer Science at Purdue University and the director of the Purdue Cyber Center. Dr. Elmagarmid served as Director of both the Indiana Center for Database Systems and the Cyber Center at Purdue University’s Discovery Park. Dr. Elmagarmid holds a Ph.D. degree in Computer Science from Ohio State University. For more information...|
|Ernesto Damiani is currently Director of the Information Security Research Center at Khalifa University of Technology, Abu Dhabi. He is the leader of the Big Data Initiative at the Etisalat British Telecom Innovation Center (EBTIC). Ernesto is on extended leave from the Department of Computer Science, Università degli Studi di Milano, Italy, where he leads the SESAR research lab (http://sesar.di.unimi.it) and has served the Head of the Ph.D. Program in Computer Science for many years . Ernesto's research interests include secure cloud-based architectures, privacy-preserving Big Data analytics and Cyber-Physical Systems security. For more information...|
|Salim Hariri is a Professor in the Department of Electrical and Computer Engineering at The University of Arizona. He received his
Ph.D. in computer engineering from University of Southern California in 1986, MSc from The Ohio State University in 1982 and B. S. from Damascus University in 1977. He is the UA site director of NSF Center for Cloud and Autonomic Computing and he is the
Editor-In-Chief for the CLUSTER COMPUTING JOURNAL (Springer, http://clus.edmgr.com). For more information...
Breaking Big Data Silos: Opportunities and Threats
Talk abstract (Ernesto Damiani):
It is largely acknowledged that organizations can unlock service and product innovation by using Big Data Analytics (BDA).
Today, companies and government agencies alike have access to more customer information than ever before. Transactional data can provide deep insight into individual customers’ behavior, to be complemented with user profiles, website clickstreams, call center logs and emails, and external data coming from social media. Experience has shown that customer acquisition (attracting new customers), customer life-time value maximization (cross-selling and up-selling to existing customers) and churn prevention (preventing customers from switching to another provider) are expected to work best when tailored to individual customers’ needs. BDA also promises to find innovative ways to reduce risk., e.g. addressing fraud threats without antagonizing regular customers.
The overall recipe advertised by vendors for enabling BDA-driven innovation is breaking down the IT “silos”, bringing data from different departments together into a common representation and a single source of knowledge.
However, adopting a “silos-breaking” data fusion strategy can itself be a source of risk and regulatory compliance headaches. Pulling out and merging data from various sources for analytics may increase the numer of replicas of sensitive data items, and increase the likelihood of data leakages or inference attacks. Also, it may make it difficult to avoid violations of custody obligations and usage constraints on the original data.
The talk will provide a threat/opportunity analysis of data preparation and fusion strategies, highlighting the research issues to be handled in order to enable a risk-aware approach to data-driven innovation.