Tutorial 1
Graph Mining Approaches: From Main Memory to Map/Reduce
Sharma Chakravarthy
Practitioners and professionals requiring up-to-date information on latest trends in newer forms of mining paradigms and how to apply these techniques for various applications, such as dealing with very large graph sizes, partitioning techniques, graph query answering etc. will benefit from this tutorial. The presenter has been working for over a decade on graph mining, scalability issues of graph mining, and its applications. Although graph mining itself has been around for a long while, it has come to the forefront due to its ability to make a difference in such domains as fraud monitoring and more recently analyzing very large social networks. Conventional mining techniques do not lend themselves to some of these applications as they cannot represent inherent structural relationships and exploit them during mining. We will present several graph mining approaches that have been proposed in the literature and new ones that are being developed. Practitioners will benefit from the practical nature of the topics and find the solutions presented applicable to problems they have encountered. Researchers will benefit from the issues that need to be addressed in one of the hot areas currently being revolutionized by increasing amounts of information available using large computing farms.