Information Acquisition is the most essential and the most widely¡Vused function on the Internet. The Internet provides a platform on which News to be communicated to users faster and services to be delivered more efficiently. With the growth of users on the Internet, the online news database becomes a powerful tool of information acquisition under the integration of computers and communication technologies. Users can find useful information related to their work and daily life very easily. The purpose of this research is to investigate the information acquisition behavior affiliated with different uses of on-line news database. We use the Task-Technology Fit Model proposed by Goodhue and Thompson (1995) to investigate whether a better fit can result in a better decision satisfaction. An experiment was designed to study the user satisfaction under different settings of task characteristics and search tools. The results indicate that different search tools result in different satisfaction levels. More specifically, the keyword search has resulted a higher user satisfaction than the classified catalog search tool. Unfortunately, we do not find a better fit between task characteristics and the search tool can result in a higher user satisfaction. The findings can tell why Google, a pure keyword search engine, becomes dominating on the Internet search. It also provides implications for the design of search tools and the improvement on the functions of classified catalogs for developing on-line news Databases.

Práce pojednává o využití support vector machines v detekci řečové aktivity. V první části jsou zkoumány různé druhy příznaků, jejich extrakce a zpracování a je nalezena jejich optimální kombinace, která podává nejlepší výsledky. Druhá část představuje samotný systém pro detekci řečové aktivity a ladění jeho parametrů. Nakonec jsou výsledky porovnány s dvěma dalšími systémy, založenými na odlišných principech. Pro testování a ladění byla použita ERT broadcast news databáze. Porovnání mezi systémy bylo pak provedeno na databázi z NIST06 Rich Test Evaluations.