A Method for Generating Aggregated Associations in Databases
Oleg Romanenko graduated from Kharkiv National University of Radioelectronics (KNURE), Ukraine and has recently passed his MPhil viva at WIT with a thesis on “A Method for Generating Aggregated Associations in Databases”. The external examiner was Dr Alessandro Zanasi from Italy, and the internal examiner was Dr Andres Peratta.
Grateful Acknowledgement is given to the Foreign Office and Commonwealth Office for supporting the programme.
In this thesis, the method for generating aggregated associations is described and the major algorithms for finding association rules are discussed. Comparison of the main methods for generating association rules was carried out and detailed reviews of the method that allow aggregation associations to be generated and improvements, such as obtaining aggregated associations without candidacy generations, are presented. A program that implements this method on the computer was created. It is clear from the thesis that the association rules are frequently used in modern life and these modern techniques grow significantly behind other Data Mining tools. This method gives a possibility to create types of rules that cannot be obtained in other ways – aggregated rules give analysts access to a wider variety of information. One of the tools which has been developed enables different sources of data to be presented. This tool enables intermediate results to be displayed such as tree creation and the time of execution of the different parts of the algorithm, and also to provide statistical information for testing the method. In the final part of the thesis a number of problems are discussed for which solutions need to be found. This work provides a suitable starting point for future researchers.
As a result of the presentation and defence of the thesis both examiners recommended that he be awarded the degree of Master of Philosophy.