Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Multiclassifier systems applied to the computer-aided sequential medical diagnosis

Kurzyński M., Majak M., Żołnierek A.

Biocybernetics and Biomedical Engineering

2016

Vol. 36, no. 4

619--625

The diagnosis of patient's state based on results of successive examinations is common task in the medicine. In computer-aided algorithms taking into account the patient's history in order to improve the quality of classification seems to be very reasonable solution. In this study, two original multiclassifier systems (MC) for the computer-aided sequential diagnosis are developed, which differ with decision scheme and the methods of combining of base classifiers. The first MC system is based on dynamic ensemble selection scheme and works in two-level structure. The second MC system in combining procedure uses original concept of meta-Bayes classifier and produces decision according to the Bayes rule. Both MC systems were practically applied to the diagnosis of human acid–base equilibrium states and compared with some state-of-the-art sequential diagnosis methods. Results obtained in experimental investigations imply that MC system is effective approach, which improves recognition accuracy in sequential diagnosis scheme.

Pipelined language model construction for Polish speech recognition

Sas J., Żołnierek A.

International Journal of Applied Mathematics and Computer Science

2013

Vol. 23, no. 3

649--668

The aim of works described in this article is to elaborate and experimentally evaluate a consistent method of Language Model (LM) construction for the sake of Polish speech recognition. In the proposed method we tried to take into account the features and specific problems experienced in practical applications of speech recognition in the Polish language, reach inflection, a loose word order and the tendency for short word deletion. The LM is created in five stages. Each successive stage takes the model prepared at the previous stage and modifies or extends it so as to improve its properties. At the first stage, typical methods of LM smoothing are used to create the initial model. Four most frequently used methods of LM construction are here. At the second stage the model is extended in order to take into account words indirectly co-occurring in the corpus. At the next stage, LM modifications are aimed at reduction of short word deletion errors, which occur frequently in Polish speech recognition. The fourth stage extends the model by insertion of words that were not observed in the corpus. Finally the model is modified so as to assure highly accurate recognition of very important utterances. The performance of the methods applied is tested in four language domains.