|
|
Behaviour-Based Identification of Network Devices
Polák, Michael Adam ; Holkovič, Martin (referee) ; Polčák, Libor (advisor)
Táto práca sa zaoberá problematikou identifikácie sieťových zariadení na základe ich chovania v sieti. S neustále sa zvyšujúcim počtom zariadení na sieti je neustále dôležitejšia schopnosť identifikovať zariadenia z bezpečnostných dôvodov. Táto práca ďalej pojednáva o základoch počítačových sietí a metódach, ktoré boli využívané v minulosti na identifikáciu sieťových zariadení. Následne sú popísané algoritmy využívané v strojovom učení a taktiež sú popísané ich výhody i nevýhody. Nakoniec, táto práca otestuje dva tradičné algorithmy strojového učenia a navrhuje dva nové prístupy na identifikáciu sieťových zariadení. Výsledný navrhovaný algoritmus v tejto práci dosahuje 89% presnosť identifikácii sieťových zariadení na reálnej dátovej sade s viac ako 10000 zariadeniami.
|
|
|
Comparison of Classification Methods
Dočekal, Martin ; Zendulka, Jaroslav (referee) ; Burgetová, Ivana (advisor)
This thesis deals with a comparison of classification methods. At first, these classification methods based on machine learning are described, then a classifier comparison system is designed and implemented. This thesis also describes some classification tasks and datasets on which the designed system will be tested. The evaluation of classification tasks is done according to standard metrics. In this thesis is presented design and implementation of a classifier that is based on the principle of evolutionary algorithms.
|
|
|
Analysis of Classification Methods
Juríček, Jakub ; Zendulka, Jaroslav (referee) ; Burgetová, Ivana (advisor)
This work deals with the classification methods used in the knowledge discovery from data process and discusses the possibilities of their validation and comparison. Through experiments, the work focuses on the analysis of four selected methods: Naive Bayes classificator, decision tree, neural network and SVM. Factors influencing basic characteristics such as training speed, classification speed, accuracy are examined. A part of the thesis is a desktop application, which is a tool for training, testing and validation of individual methods. Eleven reference data sets are selected for experimental purposes. At the end of this work experimental results of comparison and observed characteristics of classification methods are summarized.
|
|
|
Decision making based on partially known decision trees
Poláček, Tomáš ; Dostál, Petr (referee) ; Koutský, Jaroslav (referee) ; Váchal, Jan (referee) ; Dohnal, Mirko (advisor)
There is a wide range of different algorithms for insolvency prediction. The complex concept of insolvency proceedings from the point of view of both parties (debtor versus creditor) and from the point of view of the macroeconomics in this dissertation is new. It is often very difficult to generate forecasts using numerical quantifiers and traditional statistical methods. The reason is the lack of input data. Therefore, the work uses trend analysis tools based on the least information intensive quantifiers, ie trends, increasing, constant, and decreasing. A trend model solution is a set of scenarios where a set of variables is quantified by these trends. All possible transitions between the scenarios are generated and plotted in transition graphs. The oriented transition graph has as a node a set of scenarios, and as a branch the transitions between the scenarios. The given path through the transition graph describes any possible future and past behavior of the insolvency system being investigated. The Transition graph is a complete list of trend-based forecasts. The heuristics for determination of the payoff values from the insolvency proceedings applicable to the decision tree tools and the generated transition graphs from trend analyzes are also presented and used in the thesis. A nine-dimensional model serves as a case study. Vague variables are used in models that may have a major impact on the entire insolvency process, eg greed level and political situation.
|
|
|
Machine Learning Optimization of KPI Prediction
Haris, Daniel ; Burget, Radek (referee) ; Bartík, Vladimír (advisor)
This thesis aims to optimize the machine learning algorithms for predicting KPI metrics for an organization. The organization is predicting whether projects meet planned deadlines of the last phase of development process using machine learning. The work focuses on the analysis of prediction models and sets the goal of selecting new candidate models for the prediction system. We have implemented a system that automatically selects the best feature variables for learning. Trained models were evaluated by several performance metrics and the best candidates were chosen for the prediction. Candidate models achieved higher accuracy, which means, that the prediction system provides more reliable responses. We suggested other improvements that could increase the accuracy of the forecast.
|
|
|
Data Mining for Suggesting Further Actions
Veselovský, Martin ; Burget, Radek (referee) ; Bartík, Vladimír (advisor)
Knowledge discovery from databases is a complex issue involving integration, data preparation, data mining using machine learning methods and visualization of results. The thesis deals with the whole process of knowledge discovery, especially with the issue of data warehousing, where it offers the design and implementation of a specific data warehouse for the company ROI Hunter, a.s. In the field of data mining, the work focuses on the classification and forecasting of the advertising data available from the prepared data warehouse and, in particular, on the decision tree classification. When predicting the development of new ads, emphasis is put on the rationale for the prediction as well as the proposal to adjust the ad settings so that the prediction ends positively and, with a certain likelihood, the ads actually get better results.
|
|
|
The use of decision trees in managerial practice
Lapiankova, Hanna ; Váchová, Lucie (advisor) ; Zelená, Veronika (referee)
The aim of this thesis is to present the characteristics of decision trees and their use in practice and to describe the solution of decision trees with the use of PrecisionTree software. The work is divided into theoretical and practical parts. In the theoretical part the basic information about decision trees and their software support is described. In the practical part the use of decision trees for managerial decision-making with help of software support is shown on example.
|
|
|
Adaptive Similarity of XML Data
Jílková, Eva ; Holubová, Irena (advisor) ; Svoboda, Martin (referee)
In the present work we explore application of XML schema mapping in conceptual modeling of XML schemas. We expand upon the previous efforts to map XML schemas to PIM schema via a decision tree. In this thesis more versatile method is implemented - the decision tree is trained from a large set of user- annotated mapping decision samples. Several variations of training that could improve the mapping results are proposed. The approach is evaluated in a wide range of experiments that show the advantages and disadvantages of the proposed variations of training. The work also contains a survey of different approaches to schema mapping and description of schema used in this work. Powered by TCPDF (www.tcpdf.org)
|
|
|
Sleep stage classification
Lacinová, Michaela ; Smital, Lukáš (referee) ; Králík, Martin (advisor)
This bachelor thesis deals with analysis of polysomnography and its methods of measurement in electroencephalography, electromyography and electrooculography in the first part. It comprises an analysis of sleep stages recommended by the AASM. Polysomnographic data are further analysed in the domains of time and frequency, which are evaluated separately. In the second part the data are classified into particular classes using methods of decision trees and k-nearest neighbours in the MATLAB programming environment. These data are evaluated and compared with available literature.
|
|
|
Semantic Recognition of Comments on the Web
Stříteský, Radek ; Harár, Pavol (referee) ; Povoda, Lukáš (advisor)
The main goal of this paper is the identification of comments on internet websites. The theoretical part is focused on artificial intelligence, mainly classifiers are described there. The practical part deals with creation of training database, which is formed by using generators of features. A generated feature might be for example a title of the HTML element where the comment is. The training database is created by input of classifiers. The result of this paper is testing classifiers in the RapidMiner program.
|
guest ::
