|
| |
|
|
Word prediction using language models
Koutný, Michal ; Popel, Martin (advisor) ; Novák, Michal (referee)
The thesis utilizes ngram language models to improve text entry with QWERTY keyboard by the means of word prediction. Related solutions are briedly introduced. Then follows theoretical background for the work. The analysis in the next part divides problems into four tasks: language model training, incorporating model for word prediction, GUI component and evaluation framework. The realization combines Python and C++. The used corpora come from Czech (19\,M words) and (84\,M words) English Wikipedia articles. A small corpus of Czech educative texts was used to test domain adaptation. The quality metrics are defined and various configuration are measured. The best solutions reduced keystrokes per character to 0.44, resp. 0.55 for English, resp. Czech on testing data.
|
|
|
Transparent Restarts of Stateless Linux Services
Šraier, Václav ; Koutný, Michal (advisor) ; Horký, Vojtěch (referee)
Services running on servers have to be updated once in a while with security patches and new features. Full-service restarts provide a simple way to update the running code, but they usually lead to costly downtime. That can, however, be avoided by preserving network connections and internal state across restarts. There is currently no general way how to do that easily. We introduce an easy-to-use library for Linux which handles transparent restarts for stateless services with minimal code requirements. We demonstrate that support from the system's service manager can significantly improve correctness. We provide an implementation of such integration into systemd, popular Linux service manager and init system. 1
|
|
|
System daemon for HelenOS
Koutný, Michal ; Děcký, Martin (advisor) ; Yaghob, Jakub (referee)
HelenOS is an operating system based on a number of cooperating server processes, however it is missing unified means to control and monitor them. The thesis first surveys approaches taken by both popular and microkernel operating systems to the service management. Further it gives a detailed overview of relevant HelenOS environment and continues by the analysis of particular service management issues. The presented knowledge is then applied to HelenOS and the implementation based on the systemd terminology is described. Finally, the thesis shortly assesses the implementation and outlines further ideas that the implementation enabled. Powered by TCPDF (www.tcpdf.org)
|
|
|
Word prediction using language models
Koutný, Michal ; Popel, Martin (advisor) ; Novák, Michal (referee)
The thesis utilizes ngram language models to improve text entry with QWERTY keyboard by the means of word prediction. Related solutions are briedly introduced. Then follows theoretical background for the work. The analysis in the next part divides problems into four tasks: language model training, incorporating model for word prediction, GUI component and evaluation framework. The realization combines Python and C++. The used corpora come from Czech (19\,M words) and (84\,M words) English Wikipedia articles. A small corpus of Czech educative texts was used to test domain adaptation. The quality metrics are defined and various configuration are measured. The best solutions reduced keystrokes per character to 0.44, resp. 0.55 for English, resp. Czech on testing data.
|
|
| |
guest ::
RSS feed.