|
|
Analysis of DDos data with clustering
Krátký, Matěj ; Šišmiš, Lukáš (referee) ; Setinský, Jiří (advisor)
This thesis focuses on the detection of distributed denial of service (DDoS) attacks using clustering algorithms. In the first part, different types of DDoS attacks and approaches to identify them are described. Next, the thesis studies clustering methods, specifically hierarchical and k-means clustering, for analyzing the network traffic associated with these attacks. It also includes the design of a detection system suitable for detecting DDoS attacks. This is followed by a description of the implementation of this system required for the analysis phase. The main part of the work consists of performing experiments on the available dataset and evaluating the effectiveness of the methods, parameters and attributes combinations used. Finally, the thesis discusses the application of the findings and the possibilities for further research in this area.
|
|
|
Comparing of Clustering Algorithms
Jakšík, Aleš ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
This bachelor thesis deals with the comparison of K-means, K-medoids and Fuzzy C-means algorithms. In the first part, it discusses the types of distance measurements, clustering evaluation options and categorization of clustering methods. In the second part, it presents a web application for comparing the selected methods and in the third part it discusses the comparison and evaluation of the results and statistics from running the clustering algorithms in the demonstration application.
|
|
|
Ischemic thrombus analysis in multiphasic brain stroke CT data
Mikešová, Tereza ; Holeček, Tomáš (referee) ; Jakubíček, Roman (advisor)
This master thesis deals with analysis of ischemic thrombus in brain CT scans. In theoretical part, a review of methods, especially thrombus segmentation, is developed. Furthermore, the anatomy of cerebral arteries and acute ischemic stroke is summarized. Selected methods from the field of image processing are briefly described. The practical part results in a comparison of thrombus segmentation methods. The segmentation itself was preceded by data preprocessing, which is described in the theses, and the creation of a manual annotation database. The best implemented method was found to be the adaptive thresholding method, which achieved a Dice score of 0,4555. By combining the methods appropriately, a final Dice score of 0,5145 was achieved. Thrombus parameters were then calculated from the segmented volumes. The median intensity value was 51,55~HU, the median length was 15,16 mm, and the median volume was determined to be 65,34 mm3. Subsequent correlation analysis showed no significant relationship between the derived parameters.
|
|
|
Knowledge Discovery from Web Logs
Valaštín, Samuel ; Rychlý, Marek (referee) ; Bartík, Vladimír (advisor)
This bachelor thesis deals with the problem of knowledge discovery from web logs. The data source in the form of web access logs allows, after appropriate preprocessing, the use of a number of techniques that are designed to deal with knowledge discovery. By applying these techniques to preprocessed data, it is possible to classify user behavior into groups, to discover interesting associations in user behavior, or to discover previously unknown sequences in common user behavior.
|
|
|
Neural Network Based Image Segmentation
Vrábelová, Pavla ; Žák, Pavel (referee) ; Švub, Miroslav (advisor)
This paper deals with application of neural networks in image segmentation. First part is an introduction to image processing and neural networks, second part describes an implementation of segmentation system and presents results of experiments. The segmentation system enables to use different types of classifiers, various image features extraction and also to evaluate the success of segmentation. Two classifiers were created - a neural network (self-organizing map) and an algorithm K-means. Colour (RGB and HSV) and texture features and their combinations were used for classification. Texture features were extracted using a set of Gabor filters. Experiments with designed classifiers and feature extractors were carried out and results were compared.
|
|
|
Object Detection and Tracking Using Interest Points
Bílý, Vojtěch ; Hradiš, Michal (referee) ; Juránek, Roman (advisor)
This paper deals with object detection and tracking using iterest points. Existing approaches are described here. Inovated method based on Generalized Hough transform and iterative Hough-space searching is proposed in this paper. Generality of proposed detector is shown in various types of objects. Object tracking is designed as frame by frame detection.
|
|
|
Unsupervised Evaluation of Speaker Recognition System
Odehnal, Ondřej ; Plchot, Oldřich (referee) ; Matějka, Pavel (advisor)
Tato práce je vystavěna nad moderním systémem pro rozpoznávání mluvčího (SID) založeného na x-vektorech. Cílem bakalářské práce je navrhnout a experimentálně vyhodnotit techniky pro evaluaci SID systému za použití audio nahrávek bez anotace tj. bez znalosti mluvčího. Pro tento účel je z každé nahrávky bez anotace vytvořen embedding. Ty se poté používají pro shlukování nahrávek a následné vytvoření pseudo-anotací. Na těchto anotacích se SID systém evaluuje pomocí equal error rate (EER) metriky. Za účelem vytvoření pseudo-anotací byly navrženy tyto shlukovací algoritmy učení bez učitele: K-means, Gaussian mixture models (GMM) a aglomerativní shlukování. Po testování vyšel jakožto nejlepší experimentální postup K-means se Silhouette metrikou, která používá kosinovou podobnost jako míru vzdálenosti. Nejlepší metoda dosáhla 5,72 % EER s referenčním EER = 5,15 %, které bylo spočítané se znalostí anotace na části datasetu SITW dev-core-core. Podobné výsledky byly získány na části datasetu SITW eval-core-core s odhadnutým EER = 5,86 % a referenčním 5,08 %. Rozdíl mezi hodnotami tvoří 0,57 % pro eval-core-core a 0, 78% pro dev-core-core. Další testy na NIST SRE16 a VoxCeleb1 datasetech byly provedeny za účelem ověření správnosti navrženého postupu. Obecně se dá říct, že navržený testovací postup měl chybu přibližně 1 %, což je poměrně dobrý výsledek pro algoritmus učení bez učitele.
|
|
|
Demonstrational Program for IZU Course
Míšová, Miroslava ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
This bachelor's thesis deals with development of new study aplications for course Fundamentals of Artificial Intelligence. These aplications are based on the older version of JavaApplet, which use features, that are no longer supported. Each applicatoin was made acording to an object-oriented paradigm and than implemented. Special care was taken in order for the UI to be intuitive and easy to use and also for the aplication to be able to be further developed.
|
|
|
Systems for remote measurement in power engineering
Hudec, Lukáš ; Mlýnek, Petr (referee) ; Mišurec, Jiří (advisor)
The work deals with the measurement and management in power. Provides an introduction to the problems of remote meter reading, management, and describes the current situation in the field of modern technologies Smart metering and Smart grids. It also analyzed the issue of collection of networks and data collection from a large number of meters over a wide area. For the purpose of data transmission are described GPRS, PLC, DSL, ... Further, there are given options to streamline communication. This area is used hierarchical aggregation. Using k-means algorithm is a program designed to calculate the number of concentrators and their location in the group of meters. The finished program is written in Java. It has a graphical interface and shows how the calculation is conducted. To verify the results of the optimization program is given simulation model in OPNET Modeler tool. Audited results are described in the conclusion and can deduce that using the optimization program is to streamline communications.
|
|
|
Knowledge Discovery in Multimedia Databases
Málik, Peter ; Bartík, Vladimír (referee) ; Chmelař, Petr (advisor)
This master"s thesis deals with the knowledge discovery in multimedia databases. It contains general principles of knowledge discovery in databases, especially methods of cluster analysis used for data mining in large and multidimensional databases are described here. The next chapter contains introduction to multimedia databases, focusing on the extraction of low level features from images and video data. The practical part is then an implementation of the methods BIRCH, DBSCAN and k-means for cluster analysis. Final part is dedicated to experiments above TRECVid 2008 dataset and description of achievements.
|
guest ::
