|
|
Semantic relation extraction from unstructured data in the business domain
Rampula, Ilana ; Pecina, Pavel (advisor) ; Kuboň, Vladislav (referee)
Text analytics in the business domain is a growing field in research and practical applications. We chose to concentrate on Relation Extraction from unstructured data which was provided by a corporate partner. Analyzing text from this domain requires a different approach, counting with irregularities and domain specific attributes. In this thesis, we present two methods for relation extraction. The Snowball system and the Distant Supervision method were both adapted for the unique data. The methods were implemented to use both structured and unstructured data from the database of the company. Keywords: Information Retrieval, Relation Extraction, Text Analytics, Distant Supervision, Snowball
|
|
|
Content-based exploration of unstructured data
Čech, Přemysl ; Lokoč, Jakub (advisor) ; Barthel, Kai Uwe (referee) ; Gudmundsson, Gylfi Thor (referee)
Effective analysis, searching and browsing throughout arbitrary multimedia collections is still a challenging task. To perform a search among multimedia objects, first, a similarity model has to be defined. Such a model establishes methods describing how the content of individual objects is processed and how key features and descriptors, that are used for modeling similarity between objects, are formed. This task is not trivial since there can be many ways of determining how to comprehend the content of multimedia data. Furthermore, with the growing size of contemporary database collections, multimedia retrieval and exploration are extremely computationally intensive. Hence, researchers investigate support indexing structures that can evaluate similarity queries and can respond to user's queries in almost real-time even on datasets counting billions of objects. Another very important aspect of a retrieval system is the user interface for defining queries as well as presenting retrieved results. A multimedia system should offer various inputs for formulating user's queries, especially for situations in which a user cannot provide an ideal query example. Finally, a well- arranged and easy to read interface for visualization of retrieved results is essential for the success of a multimedia exploration and...
|
|
|
Design and Implementation of System for Aggregations of Real Estate Offers in the Czech Republic
Drobník, Jakub ; Kučera, Jan (advisor) ; Chlapek, Dušan (referee)
The diploma thesis deals with the design and implementation of software for aggregations of real estate offers in the Czech Republic. The aim of the thesis is to create a system which aggregates the data of real estate offers from web pages. This thesis consists of two basic parts. The context of creating the system is described in the first part. The author discusses ways to retrieve data from websites - especially the extraction of data using automated robots - in the first part of the thesis. The design and implementation of the system are described in the second part. The author and sponsor define requirements for the system in the second part of the thesis. The outcome of this thesis is a prototype that aggregates data from real estate portals into the prepared database. The main contribution of the thesis is an example of a possible approach that can aggregate data from a particular market segment and put it into the database.
|
|
|
Application of text mining methods for analysis of users movie reviews
Palatínus, Vojtěch ; Matějka, Martin (advisor) ; Novotný, Ota (referee)
The topic of this thesis is to define the challenges while working with the unstructured data. It focuses, specifically, on a transformation between unstructured and structured data using text mining methods and bringing the closer view on so-called Big Data phenomenon. The goal of this thesis is to introduce problems that occur when working with unstructured data, to show their transformation to structured data format using text mining methods and to perform analysis on user reviews published on the website of The Internet Movie Database from the mined data. The aim of this thesis is to familiarize the reader with the unstructured data and on the example demonstrate how to use text mining methods for mining relevant information from this type of data.
|
|
|
Semantic relation extraction from unstructured data in the business domain
Rampula, Ilana ; Pecina, Pavel (advisor) ; Kuboň, Vladislav (referee)
Text analytics in the business domain is a growing field in research and practical applications. We chose to concentrate on Relation Extraction from unstructured data which was provided by a corporate partner. Analyzing text from this domain requires a different approach, counting with irregularities and domain specific attributes. In this thesis, we present two methods for relation extraction. The Snowball system and the Distant Supervision method were both adapted for the unique data. The methods were implemented to use both structured and unstructured data from the database of the company. Keywords: Information Retrieval, Relation Extraction, Text Analytics, Distant Supervision, Snowball
|
|
|
Usage of unstructured data in Business Intelligence
Rakhmanova, Malika ; Šperková, Lucie (advisor) ; Karkošková, Soňa (referee)
The aim of the thesis is to identify the main trends that are occurring in the market of Business Intelligence and related to unstructured data, to describe the possibilities for integrating unstructured data, to clarify what the impact on the company have the results that can be obtained using these solutions and how generally incorporate an analysis of unstructured data into BI. Another aim is to show the current situation of processing unstructured data on the example of BI system. The thesis is divided into several parts. First part is describing of the Business Intelligence area and the basic components of Business Intelligence, as well as identifying market trends. Then, there is the next part: separating the data into structured and unstructured. Here is the part about how you can access and analyse unstructured data and what is their place in BI systems. This is the end of a block of unstructured data and the beginning of a description of the enhanced version of BI. Finally, the current market situation and BI tools, which include unstructured data, are introduced. This section provides an overview of how BI tools approach to analyse unstructured data. Existed literature, professional and freely available Internet resources are used for writing the work. The purpose is to serve as a source of information for quickly orienting in the current situation, to serve as a guide to the world of BI solutions and to show potential users what are the options and functionality of these BI solutions.
|
|
|
The analyses of unstructured content from publicly available social media by Watson
Šverák, Martin ; Molnár, Zdeněk (advisor) ; Hawlová, Kateřina (referee)
This graduate thesis deals with the analysis of unstructured data from public social media. In particular, it deals with the analysis of data from social media of Vodafone Czech Republic a.s. This thesis is divided into two parts. The first part provides theoretical background for the second part. Therefore, the first part describes social media, structured and unstructured data and tools which are used for analysing of unstructured data. In the second part, tool Watson is used for the analysis of publicly available data. Then, methodology is designed to control the analysis process and subsequently this methodology used in the formation of the pilot application that has to verify the functionality of unstructured data by tool Watson. The results of the analysis are in the conclusion. The main benefits of this thesis are the development of a pilot application of Watson and the verification of its functionality. The pilot application cannot be equated with a complete analysis that can be done by Watson. But this pilot application may work as a demonstration of Watson's functionalities.
|
|
|
Competitive analysis of leading ICT companies on the Czech market
Dvořák, Oskar ; Feige, Tomáš (advisor) ; Molnár, Zdeněk (referee)
This thesis deals with the field of Competitive Intelligence in relation to the possibilities of application of its methods and tools for competitive analysis of the market environment using modern virtual social networks. Theoretical part focuses on the characteristics of the market environment of ICT companies by using Porter's analysis and then it is focused on the description of selected tools and methods used to processing unstructured data and social networks analysis. The practical part is based on a real project which ran from early March 2013 at IBM Company. Practical part demonstrates current possibilities of information technology in the field of Competitive Intelligence.
|
|
|
Využití sociálních sítí v Competitive Intelligence
Feige, Tomáš ; Molnár, Zdeněk (advisor) ; Švík, Martin (referee)
This thesis focuses on the area of competitive intelligence with the emphasis on new possibilities and opportunities in relation to modern social networks. First it gives general analysis of the current state of competitive intelligence market as a whole and then deals with individual major leaders and their products, thus providing detailed overview of this business segment. It also discusses the possibilities of using social networks and other social or soft sources for competitive intelligence. Practical part of the thesis then demonstrates the theoretical knowledge on a real life CI project, which took place in early 2012 in cooperation with experts from IBM, including some interesting results and findings in appendix. The whole chapter can be used as a reference model for future projects with similar goals.
|
|
|
Storing hierarchical and unstructured data with Java Content Repository
Pytelka, Petr ; Pavlíčková, Jarmila (advisor) ; Feuerlicht, Jiří (referee)
This paper discusses the possibilities of storing hierarchical and unstructured data using standards JSR-170 and JSR-283 - "Content Repository for Java". Background of this paper is the graph theory. A definition of hierarchical data that is based on this theory is presented in the paper. Other methods of storing data such as the file-system, the database systems and the content management systems are discussed. The paper provides a detailed description of standard JSR-283 itself and the available features thereof. This is followed by a comparison of relation-, object-relational databases and the features of the individual techniques of object-relational mapping. Reference implementation JackRabbit is described in detail. It includes the description of the relevant API and its configuration. A case study dealing with the realization of the internal structure of a document management system is a part of this paper. Some performance tests were carried out on the reference implementation; the results thereof are presented in the paper. The conclusion of the work provides for a set of criteria to determine situations where it is appropriate to use a repository compatible with JSR-170/283 to store hierarchical and unstructured data, or where reference implementation JackRabbit can be used.
|
guest ::
