Association rule discovery has emerged as an important problem in knowledge discovery and data mining. The association mining task consists of identifying the frequent itemsets, and then forming conditional implication rules among them. In this paper we present efficient algorithms

of data management problems, such as data integration, query optimization, and the maintenance of physical data independence. To date, the performance of proposed algorithms has received very little attention, and in particular, their scale up in the presence of a large number of views is unknown. We

new scalable algorithm for spin locks that generates O(1) remote references per lock acquisition, independent of the number of processors attempting to acquire the lock. Our algorithm provides reasonable latency in the absence of contention, requires only a constant amount of space per lock

Considering the scalability of using formal concept analysis to locate features in source code, we present a set of alternative straightforward algorithms to achieve the same objectives. A preliminary experiment indicates that the alternative algorithms are more scalable to deal with the large

he/she bought). We want to find similar users, e.g., for market segmentation, anomaly/fraud detection. We propose to address this problem and present D-Search, which includes fast and effective algorithms for similarity search in large distribution datasets. Our main contributions are (1) approximate

To my parents and teachers. ii ACKNOWLEDGEMENTS I am grateful for having the ideal advisor, Professor Petar Momcilovic. Working with Professor Momcilovic has brought me the greatest joy of learning. His industrious and rigorous attitudes made him the paragon of a researcher throughout my graduate study. Without his vision and guidance, this dissertation simply would not have been possible. I would like to express my gratitude towards my dissertation committee. I thank Professor Demosthenis Tenekitzis for his comprehensive and valuable comments on my research. He has also been an inspiring teacher in my mathematical coursework. I am grateful to Professor Mingyan Liu for her attentive reviews on my thesis draft and her excellent teaching in my very first course on communication in Michigan. I also thank Professor Mark Van Oyen whose course intrigued my interest in queueing theory. My gratitude extends to all my teachers in the Electrical Engineering: Systems, Mathematics, Statistics, Computer Science and Industrial & Operations Engineering Departments. In particular, I thank Professor Serap Savari for her warm and helpful advices. I am grateful for having the company of my classmates and colleagues. I thank my friends, for being consistently supportive and encouraging. They have made my graduate life cheerful and enjoyable. This dissertation would not be complete without the mention of my respectful par-ents. Their dedication to my education is the primary reason for my love towards science and engineering. I sincerely thank them for their confidence in my capability of doing anything with determination. iii

, and integration methods along with the efficient electrostatics evaluation algorithms employed and temperature and pressure controls used. Features for steering the simulation across barriers and for calculating both alchemical and conformational free energy differences are presented. The motivations for and a

We survey techniques in the Toolkit for Advanced Optimization (TAO) for developing scalable algorithms for mesh-based optimization problems on distributed architectures. We discuss the distribution of the mesh, the computation of the gradient and the Hessian matrix, and the use of preconditioners

importantly, our techniques also extend to the general setting of non-uniform page-sizes and dependent-requests. This is the first scalable algorithm for broadcast scheduling with varying size We consider the classic problem of scheduling in a broadcast setting to minimize the average response time

scalability together with a random access property. The algorithm has modest complexity and is extremely well suited to applications involving remote browsing of large compressed images. The algorithm lends itself to explicit optimization with respect to MSE as well as more realistic psychovisua1 metrics