Evaluation of Expressions with Uncertainty in Databases

iii Acknowledgement I wish to express my gratitude to a number of people who became involved with this thesis, one way or another. I am deeply indebted to my supervisor, Dr. Nauman Chaudhry whose help, stimulating discussions; suggestions and encouragement helped me in all the time of research and for writing of this thesis. In addition, I would like to thank Dr. Mahdi Abdelguerfi and Dr. Golden Richard III for being on my thesis defense committee. Finally, I would like to thank my family and friends for the continuous support and help they provided to me. I would like to dedicate this thesis to my parents. Abstract Expressions are used in a range of applications like Publish/Subscribe, Ecommerce, etc. Integrating support for expressions in a database management system (DBMS) provides an efficient and scalable platform for applications that use Expressions. Support from uncertain data and expressions can be beneficial but not currently provided for. In this thesis, we investigate how expressions with uncertainty can be integrated in a DBMS like other data. We describe the underlying theory and implementation of UNXS (UNcertain eXpression System), a system that we have developed to handle uncertainty in expressions and data. We develop a theoretical model to compare and contrast different previous work in supporting uncertainty in DBMS and Publish/Subscribe systems. We extend the existing approaches to propose new techniques for matching uncertain expressions to uncertain data in UNXS. We then describe an implementation that integrates this support in Postgresql DBMS, which to our knowledge is the first such implementation.. Expressions are a good way to model the interests of a user in expected data. Expressions have been used in various application domains, such as, Publish/Subscribe [2, 13], Ecommerce [11], Website Personalization [12], etc. In such application domains, users want to specify their interest in expected data in terms of expressions defined on this data. The application needs to persistently maintain these expressions and match data with these expressions to inform users of items of their interest. For example, in a Publish/Subscribe system [13, 2] for a " Real Estate " application, subscriptions correspond to the interests of the user defined over various attributes of houses. These subscriptions are matched against publications of houses being sold that are published by an information provider. Subscription: Notify me of houses with price less than forty thousand that have a garage and have more than two bedrooms. …