|
|
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.
|
|
|
Data and Ontologies
Kotowski, Jakub ; Štěpánek, Petr (advisor) ; Vojtáš, Peter (referee)
In the core of current semantic web efforts is the notion of ontology. Although ontologies have been explored since times of the Greek philosopher Aristotle already there is still much to learn in this area. There are many tutorials and introductory texts both into the semantic web as whole and into its individual technologies but the learning curve is still very steep and developers have to seek answers to questions that are often topic of current research or the answer is simply not known yet. Much of these difficulties one encounters at what is often the very first step towards building a semantic web application - creating an ontology. In this work I point out some of the difficulties I encountered myself during the work on an ontology for a Sun Microsystems website. I review current research on the problems and suggest a solution. I particularly analyze the notion of role concepts, suggest an improvment of a normalisation technique which can then be used to avoid some of the role-related problems. I also present the process of creating the example Sun ontology.
|
|
| |
|
|
Acquiring user preferences for eshop
Smrčka, Zdeněk ; Vojtáš, Peter (advisor) ; Eckhardt, Alan (referee)
Title: Acquiring user preferences for e-shop Author: Zdenek Smrcka Department: The Department of Software Engineering Supervisor: prof. RNDr. Peter Vojtáš, DrSc. Abstract: The goal of this thesis is to create e-shop /in medical domain (sector), but useable in other domains (sectors)/. We use implicit metod for acquiring user preferences. By the help of this method the products are ordered from most preferred to less preferred for registered user. In case that user will choose some interesting category of products, then products are listed from maximum to less preferred in that category. It creates feedback about popularity of products for individual users and the producer gets information about interest of products in the market place and administrator can change positions of products in e-shop base on popularity. Keywords: user preferences, internet shop, user's favorite's products
|
|
|
Theoretical aspect of modeling of user decision
Lacký, Peter ; Vojtáš, Peter (advisor) ; Vomlelová, Marta (referee)
In this thesis we address to the problematics of modelling user preferences. We discuss different views on user preferences as well as we give an overview of known models of user preferences and compare them. In more detail we introduce Fuzzy Logic Programming, Bayesian Logic Programming, Probabilistic Relational Models and Markov Logic Networks. For each model we propose transformations to other models and we show possible utilizations in real world. Finally we present our suggestions how to extend and improve these models. Powered by TCPDF (www.tcpdf.org)
|
|
|
Content based Recommendation from Explicit Ratings
Ferenc, Matej ; Vojtáš, Peter (advisor) ; Peška, Ladislav (referee)
In the thesis we compare several models for prediction of user preferences. The focus is mainly on Content Based models which work with metadata about objects that are recommended. These models are compared with other models which do not use metadata for recommendation. We use three datasets and three metrics to get the results of recommendation. The goal of the thesis is to find out how can the metadata about the users and the objects enhance the standard recommender models. However, the result is that the metadata can enhance recommendation in some cases, but it varies by used metrics and dataset. This enhancement is not significant.
|
|
|
User preferences in the domain of web shops
Peška, Ladislav ; Vojtáš, Peter (advisor) ; Eckhardt, Alan (referee)
The goal of the thesis is first to find available information about user preferences, user feedback and their acquisition, processing, storing etc. The collected information is then used for making suggestions / advices for the creating an recommender system for the web shops (with special emphasis on implicit feedback). The following chapters introduces UPComp - our solution of the recommender system for the web shops. The UPComp is written in the programming language PHP and uses MySQL database. The thesis also includes testing of the UPComp on real-user web shop sites slantour.cz and antikvariat-ichtys.cz.
|
|
|
Models of user preferences in e-shop environment
Václav, Branislav ; Vojtáš, Peter (advisor) ; Ondreička, Matúš (referee)
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.
|
|
|
Modelling n-ary relations in description logics
Blaško, Miroslav ; Zavoral, Filip (advisor) ; Vojtáš, Peter (referee)
DLR is an expressive description logic with support of n-ary relations. Currently, there is no known algorithm for native reasoning within DLR. However there are two approaches that allow to delegate reasoning services of DLR to binary description logics. In this work we de ne new description logic NDL, a subset of DLR, for which we believe that native reasoning can be provided. Based on the existing approaches, we transform NDL to binary description logics for which the current of state-of-art of reasoners exist. New transformations will be analysed both theoretically and empirically. N-ary data for benchmark will be created from existing OWL ontologies by transformation of opposite direction. This benchmark can be used for comparison of native reasoning and reasoning by transformation to binary DLs.
|
|
|
Univerzální doporučovací systém
Cvengroš, Petr ; Vojtáš, Peter (advisor) ; Dědek, Jan (referee)
Recommender systems are programs that aim to present items like songs or books that are likely to be interesting for a user. These systems have become increasingly popular and are intensively studied by research groups all over the world. In web systems, like e-shops or community servers there are usually multiple data sources we can use for recommending, as user and item attributes, user-item rating or implicit feedback from user behaviour. In the thesis, we present a concept of a Universal Recommender System (Unresyst) that can use these data sources and is domain-independent at the same time. We propose how Unresyst can be used. From the contemporary methods of recommending, we choose a knowledge based algorithm combined with collaborative filtering as the most appropriate algorithm for Unresyst. We analyze data sources in various systems and generalize them to be domain-independent. We design the architecture of Unresyst, describe its interfaces and methods for processing the data sources. We adapt Unresyst to three real-world data sets, evaluate the recommendation accuracy results and compare them to a contemporary collaborative filtering recommender. The comparison shows that combining multiple data sources can improve the accuracy of collaborative filtering algorithms and can be used in systems where...
|
guest ::
