|
|
Brno Communication Agent
Jurkovič, Juraj ; Fajčík, Martin (referee) ; Smrž, Pavel (advisor)
The goal of this thesis is explore and subsequently apply techniques and technical solutions in development of information agents. Thesis primarily focuses on solving individual sub tasks using state of the art systems, interconnecting these systems, their adoption for specific domain and implementation of individual modules of communication agent system. User interface is based on multi-platform chat application Telegram. Information extraction from user input is executed by Dialogflow. Several external services are used for user request fulfillment. Elasticsearch is used for searching structured data. For answering open domain questions from free text we use R-net implementation. The resulting can have both ,its knowledge base and range of requests it can fulfill, easily extended and can be deployed to chat platform of choice.
|
|
|
Stylized Natural Language Generation in Dialogue Systems
Bolshakova, Ksenia ; Kesiraju, Santosh (referee) ; Fajčík, Martin (advisor)
Tato práce se zabývá přístupy generování přirozeného jazyka v různých stylech. Kromě toho také zkoumá schopnost modelů řídit sílu projevu stylu v generované sekvenci. Model pro generování přirozeného jazyka byl implementován s několika aspekty projevů stylu, konkrétně poezie, humor, sentiment a specifičnost. Jako strategie dekódování jazykových modelů byly použity Beam search a Nucleus sampling. Navrhované experimenty jsou založeny na váženém dekódování. Zejména pravděpodobnostní funkce vypočítaná pomocí jazykového modelu, který generuje odpověď, je modifikována dvěma přístupy. První přístup používá ručně vytvořené příznaky, například NIDF. Druhý používá neurální pravděpodobnostní jazykové modely natrénované na stylistických datových sadách. Architektura modelu je prezentována ve dvou verzích. První variantou je model založený na LSTM a druhá varianta využívá nejmodernější předpřipravené modely BART a GPT-2 pro generování textu. Experimenty odhalily problém, že i současné nejmodernější modely trpí špatným odhadem kompromisu mezi stylem a kontextem. Jinými slovy, čím více se styl projeví v generované sekvenci, tím méně se vztahuje k tématu diskutovanému v dialogu.
|
|
|
Novel Methods for Natural Language Generation in Spoken Dialogue Systems
Dušek, Ondřej ; Jurčíček, Filip (advisor) ; Ircing, Pavel (referee) ; Žabokrtský, Zdeněk (referee)
Title: Novel Methods for Natural Language Generation in Spoken Dialogue Systems Author: Ondřej Dušek Department: Institute of Formal and Applied Linguistics Supervisor: Ing. Mgr. Filip Jurčíček, Ph.D., Institute of Formal and Applied Linguistics Abstract: This thesis explores novel approaches to natural language generation (NLG) in spoken dialogue systems (i.e., generating system responses to be presented the user), aiming at simplifying adaptivity of NLG in three respects: domain portability, language portability, and user-adaptive outputs. Our generators improve over state-of-the-art in all of them: First, our gen- erators, which are based on statistical methods (A* search with perceptron ranking and sequence-to-sequence recurrent neural network architectures), can be trained on data without fine-grained semantic alignments, thus simplifying the process of retraining the generator for a new domain in comparison to previous approaches. Second, we enhance the neural-network-based gener- ator so that it takes preceding dialogue context into account (i.e., user's way of speaking), thus producing user-adaptive outputs. Third, we evaluate sev- eral extensions to the neural-network-based generator designed for producing output in morphologically rich languages, showing improvements in Czech generation. In...
|
|
|
Stylized Natural Language Generation in Dialogue Systems
Bolshakova, Ksenia ; Kesiraju, Santosh (referee) ; Fajčík, Martin (advisor)
Tato práce se zabývá přístupy generování přirozeného jazyka v různých stylech. Kromě toho také zkoumá schopnost modelů řídit sílu projevu stylu v generované sekvenci. Model pro generování přirozeného jazyka byl implementován s několika aspekty projevů stylu, konkrétně poezie, humor, sentiment a specifičnost. Jako strategie dekódování jazykových modelů byly použity Beam search a Nucleus sampling. Navrhované experimenty jsou založeny na váženém dekódování. Zejména pravděpodobnostní funkce vypočítaná pomocí jazykového modelu, který generuje odpověď, je modifikována dvěma přístupy. První přístup používá ručně vytvořené příznaky, například NIDF. Druhý používá neurální pravděpodobnostní jazykové modely natrénované na stylistických datových sadách. Architektura modelu je prezentována ve dvou verzích. První variantou je model založený na LSTM a druhá varianta využívá nejmodernější předpřipravené modely BART a GPT-2 pro generování textu. Experimenty odhalily problém, že i současné nejmodernější modely trpí špatným odhadem kompromisu mezi stylem a kontextem. Jinými slovy, čím více se styl projeví v generované sekvenci, tím méně se vztahuje k tématu diskutovanému v dialogu.
|
|
|
Brno Communication Agent
Jurkovič, Juraj ; Fajčík, Martin (referee) ; Smrž, Pavel (advisor)
The goal of this thesis is explore and subsequently apply techniques and technical solutions in development of information agents. Thesis primarily focuses on solving individual sub tasks using state of the art systems, interconnecting these systems, their adoption for specific domain and implementation of individual modules of communication agent system. User interface is based on multi-platform chat application Telegram. Information extraction from user input is executed by Dialogflow. Several external services are used for user request fulfillment. Elasticsearch is used for searching structured data. For answering open domain questions from free text we use R-net implementation. The resulting can have both ,its knowledge base and range of requests it can fulfill, easily extended and can be deployed to chat platform of choice.
|
|
|
Novel Methods for Natural Language Generation in Spoken Dialogue Systems
Dušek, Ondřej ; Jurčíček, Filip (advisor) ; Ircing, Pavel (referee) ; Žabokrtský, Zdeněk (referee)
Title: Novel Methods for Natural Language Generation in Spoken Dialogue Systems Author: Ondřej Dušek Department: Institute of Formal and Applied Linguistics Supervisor: Ing. Mgr. Filip Jurčíček, Ph.D., Institute of Formal and Applied Linguistics Abstract: This thesis explores novel approaches to natural language generation (NLG) in spoken dialogue systems (i.e., generating system responses to be presented the user), aiming at simplifying adaptivity of NLG in three respects: domain portability, language portability, and user-adaptive outputs. Our generators improve over state-of-the-art in all of them: First, our gen- erators, which are based on statistical methods (A* search with perceptron ranking and sequence-to-sequence recurrent neural network architectures), can be trained on data without fine-grained semantic alignments, thus simplifying the process of retraining the generator for a new domain in comparison to previous approaches. Second, we enhance the neural-network-based gener- ator so that it takes preceding dialogue context into account (i.e., user's way of speaking), thus producing user-adaptive outputs. Third, we evaluate sev- eral extensions to the neural-network-based generator designed for producing output in morphologically rich languages, showing improvements in Czech generation. In...
|
|
|
User simulation for statistical dialogue systems
Michlíková, Vendula ; Jurčíček, Filip (advisor) ; Žabokrtský, Zdeněk (referee)
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 baseline simulator, we created another bigram simulator that is trained on dialogue acts without slot values. The third implemented simulator is similar to an implemen- tation of a dialogue manager. It tracks its dialogue state and learns a dialogue strategy based on the state using supervised learning. The user simulators are implemented in Python 2.7, in ALEX framework for dialogue system development. Simulators are developed for PTICS application which operates in the domain of public transport information. Simulators are trained and evaluated using real human-machine dialogues collected with PTICS application. 1
|
|
|
Development of trainable policies for spoken dialogue systems
Le, Thanh Cong ; Jurčíček, Filip (advisor) ; Peterek, Nino (referee)
Abstract Development of trainable policies for spoken dialogue systems Thanh Le In humanhuman interaction, speech is the most natural and effective manner of communication. Spoken Dialogue Systems (SDS) have been trying to bring that high level interaction to computer systems, so with SDS, you could talk to machines rather than learn to use mouse and keyboard for performing a task. However, as inaccuracy in speech recognition and inherent ambiguity in spoken language, the dialogue state (user's desire) can never be known with certainty, and therefore, building such a SDS is not trivial. Statistical approaches have been proposed to deal with these uncertainties by maintaining a probability distribution over every possible dialogue state. Based on these distributions, the system learns how to interact with users, somehow to achieve the final goal in the most effective manner. In Reinforcement Learning (RL), the learning process is understood as optimizing a policy of choosing action conditioned on the current belief state. Since the space of dialogue...
|
|
|
Approximative Bayes methods for belief monitoring in spoken dialogue systems
Marek, David ; Jurčíček, Filip (advisor) ; Žabokrtský, Zdeněk (referee)
The most important component of virtually any dialog system is a dialogue manager. The aim of the dialog manager is to propose an action (a continuation of the dialogue) given the last dialog state. The dialog state summarises all the past user input and the system input and ideally it includes all information necessary for natural progress in the dialog. For the dialog manager to work efficiently, it is important to model the probability distribution over all dialog states as precisely as possible. It is possible that the set of dialog states will be very large, so approximative methods usually must be used. In this thesis we will discuss an implementation of approximate Bayes methods for belief state monitoring. The result is a library for dialog state monitoring in real dialog systems. 1
|
guest ::
