الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 132 |  |  |
| | | from | 132 | | |
Abstract
This study aied to : This thesis presents a framework that improves ranking processing in the case of uncertain distributed database. The proposed framework consists of three main successive layers. Three new communication and computation-efficient algorithms are investigated for retrieving the top-k tuples from distributed sites. The main concern of these algorithms is to reduce the communication rounds utilized and amount of data transmitted while achieving efficient ranking. Two of these algorithms are employed in the Ranking layer of the framework, while the third one controls the Monitoring layer. The ranking algorithms, entitled; Threshold-tuned Distributed Top-k (TDk), and Ceiling-tuned Distributed Top-k (CDk), have two versions for (attribute, and tuple) levels of uncertainty. Experimental results emphasize that both proposed algorithms have a great impact on reducing communication cost.
This thesis is organized in six chapters, as follows:
•Chapter 1 is an introduction of the thesis studying uncertain data with some examples of uncertain data and the importance of ranking queries in distributed uncertain database systems.
•Chapter 2 provides an overview of the background literature for the framework. It starts by discussing uncertain databases, highlighting uncertain database management systems with examples of current systems, uncertain data representations, uncertain database models and query processing in uncertain database.
•Chapter 3 starts with related work to the framework then discusses the problem formulation with data model definition. The framework is studied in details, highlighting the construction and the functions of each layer.
•Chapter 4 introduces the novel algorithms in attribute-level uncertain database model. The steps of each algorithm are discussed in details with a communication costs analysis for both ranking algorithms. An experimental study is performed to test the efficiency and effectiveness of the algorithms.
•Chapter 5 introduces the novel algorithms in tuple-level uncertainty model. The steps of each algorithm are discussed in details with a communication costs analysis for both ranking algorithms. An experimental study is performed to test the efficiency and effectiveness of the algorithms.
•Chapter 6 concludes the whole work and presents some ideas for future work in the same area of the thesis.