Volume 31 Issue 6 - January 5, 2018 PDF
Counter
Most Young Scholars Grant (Columbus Program) -- Privacy-preserving Social Data Mining with Its Applications
Institute of Data Science, College of Management, National Cheng Kung University
 
Font Enlarge
Online social platforms such as Facebook and Twitter allow users to interact with not only each other, but also a variety of entities via a rich set of social functions. Users can rate items, follow one another, check-in at places, and generate posts with hashtags. Although such heterogeneous social interaction data enrich user experiences and boost the profits of service providers and advertisers, they also bring privacy risk of users. The adversary can leverage artificial intelligence techniques to predict users’ visited locations based on hashtags, infer users’ rating preferences and purchasing habits, and identify the personal attributes and friends of users using the shared posts with check-in records. In this MOST Columbus Project, we aim at developing a general-purpose privacy-preserving framework based on online massive social datasets by exploiting and developing the techniques of machine learning, social network analysis, information security, and text mining. The ultimate goal is to simultaneously protect user privacy of sensitive data, and ensure data usability in various applications of data science.

The potential technique contribution of this project is three-fold, privacy-preserving machine learning algorithms, secure distributed social recommender systems, and information diffusion-based privacy prediction shielding methods. This project will also generate multifaceted profits. Users are allowed to experience secure online social services. Service providers with accurate and privacy-protected recommendation can boost traffic flow while lowering down the cost. And advertisers will raise the monetary income without leaking personal data.

The research interests of Prof. Li include data mining, machine learning, social network analysis, and recommender systems. Prof. Li would like to express his deep appreciation to the help from different units of NCKU. He sincerely thank for the support and encouragement of Departments of Statistics, Department of CSIE, and Institute of Data Science. Prof. Li expected to contribute himself for promoting NCKU’s data science in relevant international research communities, and to cultivate talents of data science and artificial intelligence to serve the society.
< Previous
Next >
Copyright National Cheng Kung University