Discussion of a master's thesis in the Faculty of Computer Science and Mathematics - Department of Statistics and Informatics
In continuation of the scientific research movement, in the presence and follow-up of the Dean of the College of Computer Science and Mathematics, Prof. Dr. Duha Bashir Abdullah, the respected
On Tuesday, 21-9-2021, the College of Computer Science and Mathematics at the University of Mosul discussed the Master's thesis of student Amal Mahmoud Naji Ahmed and under the supervision of Prof. Dr. Zakaria Yahya Al-Jammal entitled (Improving Fragmentary Clustering Using Algorithms Inspired by Nature).

The thesis dealt with the fractional clustering algorithm K-means, one of the most widely used algorithms because of its speed and accuracy in determining the number of clusters. The Monte-Carlo method was used in the simulation to generate data and estimate the optimal number of clusters.

The thesis dealt with showing the simulation results based on the total variance within the cluster as a criterion for comparison. The proposed method is superior to the methods of choosing the number of other clusters. In addition, the proposed method was applied to real data on the financial statistics of the banking system structure for the years 2016 and 2018, and the results we obtained matched The results of the simulation experiments are also superior to the other methods of choosing the number of other clusters.

The thesis aims to employ one of the algorithms inspired by nature, the luminous fireflies algorithm, the process of optimizing the number of clusters.

The discussion committee was chaired by Assistant Professor Dr. Abdul Ali Hamad from Anbar University, with the membership of Assistant Professor Dr. Nada Nemat Salim and Assistant Professor Dr. Hiyam Abdul Majeed Hayawi, under the supervision and membership of Prof. Zakaria Yahya Al-Jammal.

 

 

Also Read