|
|
Ontology mapping and mining with user preference
Maruščák, Dušan ; Hoksza, David (referee) ; Vojtáš, Peter (advisor)
In this diploma thesis we deal with the problem of ontology mining and mapping. Ontology mining is ontology learning from instance. First step to ontology mining is instance mining. We present method for ontology instance mining from web pages that contain structured data records. We use HTML tree of the page to detect data records. We search data records only in relevant branches of HTML tree and we identify also non-contiguous records. Next we present method to identify attributes of record and their mapping into attributes of ontology concepts. We discuss also several alternative methods including OCR usage for processing visual form of web document . Results of experiments on several types of web documents show a right direction of our research.
|
|
|
Using Ontologies in GUHA Procedures
Zeman, Martin ; Vojtáš, Peter (referee) ; Ralbovský, Martin (advisor)
The goal of this master thesis is to implement some of the previously suggested ways of usage of Ontologies in Knowledge Database Discovery (KDD) process. The implementation is done in the Ferda environment, a modular environment for visual GUHA data mining. The work argues about each of the suggested ways of usage of ontologies in consideration of their successive implementation. Great part of the work is devoted to selection of the right language for representation of ontologies. Furthermore, the implementation of the selected modules in Ferda is described. The functionality of these modules is then tested on an example and there is evaluated the contribution of the modules for KDD process. At the end of the work the suggestions for future development of support of ontologies during the KDD process in Ferda DataMiner are described.
|
|
|
Models of user preferences in e-shop environment
Václav, Branislav ; Ondreička, Matúš (referee) ; Vojtáš, Peter (advisor)
The aim of this work is to gain insight into the broad range of models of user preferences inside an e-shop environment. A specific group of models will be selected from the overall described set, and an exact method of calculation for these models will be introduced. The selected models, together with a corresponding web environment design, will then be implemented into a comprehensive form of a working web application. An integral part of the application is formed by the inclusion of an appropriate set of test data. Based on these data, practical experiments will be carried out, and consequent results will be considered in the assessment of the functionality of the provided application and its potential contribution for existing e-shop users. Acquired user feedback will then be used to identify further development opportunities of the implemented application.
|
|
|
Koordinace mezi logickými agenty
Krištofič, Milutín ; Vojtáš, Peter (referee) ; Štěpánek, Petr (advisor)
In this thesis, we study the recent models for coordination and we propose a new model based on argumentation system. This system, named Argumentation BlackBoard for Coordination (ABBC), and its extension, Abstract ABBC, are based on successful argumentation approaches, such as abstract argumentation framework and defeasible logic programming. Our aim is to present necessary argumentation elements in open multi-agent system that consists of logic agents. In this work, we also describe a brief software design for the system.
|
|
| |
|
|
Visualization methods utilization for document searching by user preferences
Jenček, Petr ; Kopecký, Michal (referee) ; Vojtáš, Peter (advisor)
The aim of this work is to study possibilities of presentation of document search results according to keywords given by a user using sociomapping method. Both web search and locally stored documents search will be our interest. We'll create document clusters from such results using one of the clustering algorithms and measures of keywords similarities. These clusters will be input data for creation of direct, or indirect sociomap. Finally we'll compare results of the clustering algorithms from several points of view.
|
|
|
Preferencev querying, indexing, optimisation
Horničák, Erik ; Vojtáš, Peter (advisor) ; Ondreička, Matúš (referee)
In this thesis we discuss the issue of searching the best k objects from the multi-users point of view. Every user has his own preferences, which are represented by fuzzy functions and aggregation function. This thesis designs and implements several solutions of searching the best k objects when attributes data are stored on remote servers. It was necessary to modificate existing algorithms for this type of obtaining data. This thesis uses several variants of Fagin algorithm, indexing methods using B+ trees and communication via web services.
|
|
|
Interactive Gantt Charts
Skalický, Tomáš ; Vojtáš, Peter (referee) ; Barták, Roman (advisor)
This work deals with problems of interactive rescheduling in Gantt charts. The goal of the work is to design and implement an algorithm which resolves the inconsistencies in a schedule. First existing approaches are studied. Algorithm for solving our inconsistencies in a schedule is the most important contribution of this work. It is introduced there and there is also it's proof of correctness and finiteness. Demonstration of applicability of the algorithm follows. For this reason, program GanttViewer which is a part of this work is used. Finally, this program is described in detail.
|
|
|
Methods for finding best answer with different user preferences
Eckhardt, Alan ; Vojtáš, Peter (advisor) ; Pokorný, Jaroslav (referee)
User preferences are one of new aspects in informatics, e.g. in domain of semantic web. This work tries to cover the whole problematic of user preferences, their modeling, various aspects of generating user preferences and their evaluation. It also studies the area of group decision and finding answer that suits all members of group. Next part follows the paper of Ronald Fagin, Amnon Lotem, Moni Naor about the top-k algorithm. The new independent system for the top-k algorithm is presented, some implementations of this algorithm are tested and the influence of modules of algorithm on the speed and the number of rows are studied.
|
|
|
Dynamic Datalog
Janušek, Martin ; Vojtáš, Peter (referee) ; Kosík, Antonín (advisor)
In this study we examine possibilities of integration extensional updates into Datalog. Our research involves also possibility of query evaluation optimization by making intensional rules explicit (materialization). As a result od our approach we present new query language D-Datalog. Our proposal extends Datalog database by set of constraints and set of intensional facts (materialization of intensional database). We introduce possibility of extensional and intensional updates. By intensional updates we mean updates of set of intensional facts so user can directly state, which consequences of intensional clauses should be in materialization. Updates are not executed as soon as they are evaluated. Instead, they are collected and then applied to the database when the query evaluation is completed (non-immediate update semantics). This approach is similar to semantics of relational databases. Prototype implementation of D-Datalog in Java is part of our solution.
|
guest ::
