Отрывок: For large values of n, it is more efficient to use an algorithm based on frequency features, the computational complexity of which is of the order 𝑂(𝑁 ⋅ 𝑛) or the proposed "forward-backward" algorithm. The proposed model allows us to quantify the probability of identifying the sequence of a Markov chain in terms of the possibility of generating a given AMM. In general, to identify a set ?̂?(𝑁) = 𝑢1, 𝑢2, . . . , 𝑢𝑁, the development of the "forward-b...
Полная запись метаданных
Поле DC | Значение | Язык |
---|---|---|
dc.contributor.author | Shalagin, S.V. | - |
dc.contributor.author | Nurutdinova, A.R. | - |
dc.date.accessioned | 2020-07-29 14:32:29 | - |
dc.date.available | 2020-07-29 14:32:29 | - |
dc.date.issued | 2020 | - |
dc.identifier | Dspace\SGAU\20200729\84792 | ru |
dc.identifier.citation | Shalagin S.V. Identification of Markov sequences based on a modified “forward-backward” algorithm/ S.V. Shalagin, A.R. Nurutdinova // Информационные технологии и нанотехнологии (ИТНТ-2020). Сборник трудов по материалам VI Международной конференции и молодежной школы (г. Самара, 26-29 мая): в 4 т. / Самар. нац.-исслед. ун-т им. С. П. Королева (Самар. ун-т), Ин-т систем. обраб. изобр. РАН-фил. ФНИЦ "Кристаллография и фотоника" РАН; [под ред. В. А. Фурсова]. – Самара: Изд-во Самар. ун-та, 2020. – Том 4. Науки о данных. – 2020. – С. 28-31. | ru |
dc.identifier.uri | http://repo.ssau.ru/handle/Informacionnye-tehnologii-i-nanotehnologii/Identification-of-Markov-sequences-based-on-a-modified-“forwardbackward”-algorithm-84792 | - |
dc.description.abstract | In 1989, L.R. Rabiner proposed a solution to the problem of identifying hidden Markov models by the induction method using a modified “forward-backward” algorithm. This article proposes the use of a modification of this method to solve the problem of identifying Markov sequences for belonging with a given probability to a specific class of the automate Markov model (AMM). A model defined on the basis of stochastic matrices of the class ergodic, which determine the class of the AMM. | ru |
dc.language.iso | en_US | ru |
dc.title | Identification of Markov sequences based on a modified “forward-backward” algorithm | ru |
dc.type | Article | ru |
dc.textpart | For large values of n, it is more efficient to use an algorithm based on frequency features, the computational complexity of which is of the order 𝑂(𝑁 ⋅ 𝑛) or the proposed "forward-backward" algorithm. The proposed model allows us to quantify the probability of identifying the sequence of a Markov chain in terms of the possibility of generating a given AMM. In general, to identify a set ?̂?(𝑁) = 𝑢1, 𝑢2, . . . , 𝑢𝑁, the development of the "forward-b... | - |
Располагается в коллекциях: | Информационные технологии и нанотехнологии |
Файлы этого ресурса:
Файл | Описание | Размер | Формат | |
---|---|---|---|---|
ИТНТ-2020_том 4-28-31.pdf | 445.14 kB | Adobe PDF | Просмотреть/Открыть |
Показать базовое описание ресурса
Просмотр статистики
Поделиться:
Все ресурсы в архиве электронных ресурсов защищены авторским правом, все права сохранены.