000			01357cam a2200265 a 4500
003			EG-GiCUC
008			140826s2013 ua f m 000 0 eng d
040			_aEG-GiCUC _beng _cEG-GiCUC
041	0		_aeng
049			_aGift
097			_aM.Sc
099			_aCai01.34.M.Sc.2013.Me.E
100	0		_aMennat Allah Amer
245	1	0	_aEnhancing one - class support vector machines for unsupervised anomaly detection / _cMennatallah Amer ; Supervised Markus Goldstein , Andreas Dengel , Slim Abdennadher
260			_aCairo : _bMennat Allah Amer , _c2013
300			_a77 Leaves ; _c30cm
502			_aThesis (M.Sc.) - German University - Faculty of Postgraduate Studies and Scientific Research - Department of Computer Science and Engineering
520			_aSupport vector machines (SVMs) have been one of the most prominent machine learning techniques for the past decade. In this thesis, the effectiveness of applying SVMs for derecting outliers in an unsupervised setting is investigated. Unsupervised anomaly detection techniques operate directly on an unseen dataset, under the assumption that outhers are sparsely present in it
700	0		_aAndreas Dengel , _eSupervisor
700	0		_aMarkus Goldstein , _eSupervisor
700	0		_aSlim Abdennadher , _eSupervisor
905			_aNazla _eRevisor
905			_aSamia _eCataloger
942			_2ddc _cTH
999			_c47024 _d47024