Search
Now showing items 1-10 of 25
Framework pro extrakci informací z velkého množství jazykových dat
Framework for information extraction from the large language data sets
bachelor thesis (DEFENDED)
Advisor: Križ, Vincent
Date Issued: 2014
Date of defense: 16. 06. 2014
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Tato práce popisuje program FAFEFI sloužící k extrakci n-gramů a skip-gramů z velkého množství jazykových dat. Řeší možnosti předání vstupních dat programu, návrh datových struktur pro reprezentaci n-gramů a skip-gramů v ...
This thesis describes the FAFEFI program that focuses on n-gram and skip-gram extraction from large data sets. The thesis presents two different approaches to passing input data to the program. It also describes the design ...
This thesis describes the FAFEFI program that focuses on n-gram and skip-gram extraction from large data sets. The thesis presents two different approaches to passing input data to the program. It also describes the design ...
Automatické zpracování českých soudních rozhodnutí
Processing of Czech court decisions
diploma thesis (DEFENDED)
Advisor: Vidová Hladká, Barbora
Date Issued: 2015
Date of defense: 15. 06. 2015
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Název práce: Automatické zpracování českých soudních rozhodnutí Autor: Bohdan Maslowski Katedra / Ústav: Ústav formální a aplikované lingvistiky Vedoucí diplomové práce: Mgr. Barbora Vidová Hladká, Ph.D. Abstrakt: Cílem ...
Title: Processing of Czech court decisions Author: Bohdan Maslowski Department: Institute of Formal and Applied Linguistics Supervisor: Mgr. Barbora Vidová Hladká, Ph.D. Abstract: The objective of this thesis is a comparison ...
Title: Processing of Czech court decisions Author: Bohdan Maslowski Department: Institute of Formal and Applied Linguistics Supervisor: Mgr. Barbora Vidová Hladká, Ph.D. Abstract: The objective of this thesis is a comparison ...
Unsupervised and Semi-Supervised Multilingual Learning for Resource-Poor Languages
Unsupervised and Semi-Supervised Multilingual Learning for Resource-Poor Languages
diploma thesis (DEFENDED)
Advisor: Zeman, Daniel
Date Issued: 2012
Date of defense: 07. 09. 2012
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Pra ce se zaměřuje na neř zenou morfologickou segmentaci, jednu ze za kladn ch u loh poč tačov eho zpracov an přirozen eho jazyka. V t eto u loze je c lem rozložit slova na morf emy. Popisuji a reim- plementuji model ...
This thesis focuses on unsupervised morphological seg- mentation, the fundamental task in NLP which aims to break words into morphemes. I describe and re-implement a model proposed in Lee et al. (2011) and evaluate it on ...
This thesis focuses on unsupervised morphological seg- mentation, the fundamental task in NLP which aims to break words into morphemes. I describe and re-implement a model proposed in Lee et al. (2011) and evaluate it on ...
Sledování aktivovanosti objektů v textech
Sledování aktivovanosti objektů v textech
diploma thesis (NOT DEFENDED)
Advisor: Vidová Hladká, Barbora
Date Issued: 2014
Date of defense: 08. 09. 2014
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: V kontextu analýzy diskurzu stupeň aktivovanosti (salience) modeluje aktuální míru zapojenosti odkazovaných objektů a její vývoj v průběhu textu. Algoritmus pro určování aktivovanosti a vizualizaci jejího průběhu již byl ...
The notion of salience in the discourse analysis models how the activation of referred objects evolves in the flow of text. The salience algorithm was already defined and tested briefly in an earlier research, we present ...
The notion of salience in the discourse analysis models how the activation of referred objects evolves in the flow of text. The salience algorithm was already defined and tested briefly in an earlier research, we present ...
User simulation for statistical dialogue systems
Simulace uzivatele pro statisticke dialogove systemy
diploma thesis (DEFENDED)
Advisor: Jurčíček, Filip
Date Issued: 2015
Date of defense: 09. 09. 2015
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Cílem této práce je vytvořit a evaluovat simulátory uživatele pro statistický di- alogový stystém. Simulátory pracují na úrovni dialogových aktů. Jako baseline systém jsme implementovali bigramový simulátor. Tento simulátor ...
The purpose of this thesis is to develop and evaluate user simulators for a spoken dialogue system. Created simulators are operating on dialogue act level. We implemented a bigram simulator as a baseline system. Based on ...
The purpose of this thesis is to develop and evaluate user simulators for a spoken dialogue system. Created simulators are operating on dialogue act level. We implemented a bigram simulator as a baseline system. Based on ...
Comparison of approaches to text classification
Porovnání přístupů ke klasifikaci textu
bachelor thesis (DEFENDED)
Advisor: Hana, Jiří
Date Issued: 2019
Date of defense: 05. 09. 2019
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: The focus of this thesis is short text classification. Short text is the prevailing form of text on e-commerce and review platforms, such as Yelp, Tripadvisor or Heureka. As the popularity of the online communication is ...
Popularita osob automaticky
Popularity Meter
bachelor thesis (DEFENDED)
Advisor: Bojar, Ondřej
Date Issued: 2011
Date of defense: 07. 09. 2011
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Možnost automaticky sledovat popularitu osob v~novinách by jistě uvítaly nejen tyto osoby samotné. Počítačové zpracovávání subjektivity je sice rychle se rozvíjející podobor komputační lingvistiky, v~češtině ovšem vůbec ...
Having the possibility of automatically tracking a person's popularity in the newspapers is an idea appealing not just to those in the media spotlight. While sentiment (subjectivity) analysis is a rapidly growing subfield ...
Having the possibility of automatically tracking a person's popularity in the newspapers is an idea appealing not just to those in the media spotlight. While sentiment (subjectivity) analysis is a rapidly growing subfield ...
Detekce podezřelých anotací
Detekce podezřelých anotací
bachelor thesis (DEFENDED)
Advisor: Vidová Hladká, Barbora
Date Issued: 2011
Date of defense: 20. 06. 2011
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Tato práce popisuje způsob kontroly morfologické anotace pomocí strojového učení a představuje implementaci tohoto přístupu - aplikaci MissTagger. Procedura kontroly zahrnuje jak detekci chyb, tak jejich opravu. Tento ...
This work describes a machine learning approach for checking the part-of-speech annotation, and presents its implementation - a system called MissTagger. The checking procedure covers both error detection and error correction. ...
This work describes a machine learning approach for checking the part-of-speech annotation, and presents its implementation - a system called MissTagger. The checking procedure covers both error detection and error correction. ...
Semantic information from FrameNet and the possibility of its transfer to Czech data
Sémantická informace ze sítě FrameNet a možnosti jejího využití pro česká data
diploma thesis (DEFENDED)
Advisor: Lopatková, Markéta
Date Issued: 2016
Date of defense: 08. 09. 2016
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Tématem práce je převod anotace z databáze FrameNet do češtiny a možnost využití takto vzniklých dat pro automatické předpovídání rámců. První část tohoto úkolu, převod anotace z angličtiny do češtiny, byla provedena dvěma ...
The thesis focuses on transferring FrameNet annotation from English to Czech and the possibilities of using the resulting data for automatic frame prediction in Czech. The first part, annotation transfer, has been performed ...
The thesis focuses on transferring FrameNet annotation from English to Czech and the possibilities of using the resulting data for automatic frame prediction in Czech. The first part, annotation transfer, has been performed ...
Smoothness of Functions Learned by Neural Networks
Hladkost funkcí naučených neuronovými sítěmi
bachelor thesis (DEFENDED)
Advisor: Musil, Tomáš
Date Issued: 2020
Date of defense: 07. 07. 2020
Faculty / Institute: Matematicko-fyzikální fakulta / Faculty of Mathematics and Physics
Abstract: Modern neural networks can easily fit their training set perfectly. Surprisingly, they generalize well despite being "overfit" in this way, defying the bias-variance trade-off. A prevalent explanation is that stochastic ...