Browsing Doctoral Dissertations by Title
Previous Page
Now showing items 597-600 of 600
-
Đorđević, Radmilo (Beograd , 1979)[more][less]
-
Lučić, Blagota (Belgrade , 1985)[more][less]
-
Protić, Danijela (Beograd , 2023)[more][less]
Abstract: Anomaly detection is the recognition of suspicious computer network behavior by comparing unknown network traffic to a statistical model of normal network behavior. Binary classifiers based on supervised machine learning are good candidates for normality detection. This thesis presents five standard binary classifiers: the k-nearest neighbors, weighted k-nearest neighbors, decision trees, support vector machines and feedforward neural network. The main problem with supervised learning is that it takes a lot of data to train high-precision classifiers. To reduce the training time with minimal degradation of the accuracy of the models, a two-phase pre-processing step is performed. In the first phase, numeric attributes are selected to reduce the dataset. The second phase is a novel normalization method based on hyperbolic the tangent function and the damping strategy of the Levenberg-Marquardt algorithm. The Kyoto 2006+ dataset, the only publicly available data set of real-world network traffic intended solely for anomaly detection research in computer networks, was used to demonstrate the positive impact of such pre-processing on classifier training time and accuracy. Of all the selected classifiers, the feedforward neural network has the highest processing speed, while the weighted k-nearest neighbor model proved to be the most accurate. The assumption is that when the classifiers work concurrently, they should detect either an anomaly or normal network traffic, which occasionally is not the case, resulting in different decision about the anomaly, i.e. a conflict arises. The conflicting decision detector performs a logical exclusive OR (XOR) operation on the outputs of the classifiers. If both classifiers simultaneously detected an anomaly or recognized traffic as normal, their decision was no conflict had occurred. Otherwise a conflict is detected. The number of conflicts detected provides an opportunity for additional detection of changes in computer network behavior. URI: http://hdl.handle.net/123456789/5599 Files in this item: 1
Danijela Protic - Doktorska Disertacija.pdf ( 3.143Mb ) -
Bogdanović, Katarina (Beograd , 2025)[more][less]
Abstract: In the rst and the second chapter of dissertation we prove some new inequalities for the spectral radius, essential spectral radius, oper- ator norm, measure of non-compactness and numerical radius of Hadamard (Schur) weighted geometric means of positive kernel operators on Banach function and sequence spaces. The list of extensions and re nings of known inequalities has been expanded. Some new inequalities and equalities for the generalized and the joint spectral radius and their essential versions of Hadamard (Schur) geometric means of bounded sets of positive kernel op- erators on Banach function spaces have been proved. There are additional results in case of non-negative matrices that de ne operators on Banach sequence spaces. In the third part we present some inequalities for opera- tor monotone functions and (co)hyponormal operators and give relations of Schur multipliers to derivation like inequalities for operators. In particular, let Ψ, Φ be s.n. functions, p ⩾ 2 and φ be an operator monotone function on [0, ∞) such that φ(0) = 0. If A, B, X ∈ B(H) and A and B are strictly ac- cretive such that AX−XB ∈ CΨ(H), then also AXφ(B)−φ(A)XB ∈ CΨ(H) and ||AXφ(B) − φ(A)XB||Ψ ⩽ r φ A+A∗ 2 − A+A∗ 2 φ′ A+A∗ 2 A+A∗ 2 −1 A(AX − XB)B B+B∗ 2 −1 r φ B+B∗ 2 − B+B∗ 2 φ′ B+B∗ 2 Ψ . under any of the following conditions: (a) Both A and B are normal, (b) A is cohyponormal, B is hyponormal and at least one of them is normal, and Ψ := Φ(p)∗ , (c) A is cohyponormal, B is hyponormal and ||.||Ψ is the trace norm ||.||1. Alternative inequalities for ||.||Ψ(p) norms are also obtained. URI: http://hdl.handle.net/123456789/5780 Files in this item: 1
Katarina_Bogdanovic_disertacija.pdf ( 1.621Mb )
Previous Page
Now showing items 597-600 of 600