Zobrazeno 1 - 6
of 6
pro vyhledávání: '"Ilona I. Kapitanova"'
Autor:
Andrey I. Kapitanov, Victor Dorogov, Vladimir M. Troyanovskiy, Ilona I. Kapitanova, Ekaterina Dorogova
Publikováno v:
2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus).
Predicting the marginal efficiency of the classifier allows you to reduce the requirements for training resources, which is especially important when the amount of available data is limited. Modern trends in the development of new classification algo
Autor:
Vladimir M. Troyanovskiy, Victor Dorogov, Andrey I. Kapitanov, Ilona I. Kapitanova, Ekaterina Dorogova
Publikováno v:
2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus).
Often, when solving the task of classifying text documents, we can use only a limited amount of input data. Increasing input leads to a nonlinear increase in resources such as time and memory. There may also be cases where we have to use private / pa
Autor:
Andrey I. Kapitanov, Ekaterina Dorogova, Vladimir Ilyushechkin, Vladimir M. Troyanovskiy, Ilona I. Kapitanova
Publikováno v:
2019 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus).
When processing text documents in natural language, you may encounter such problem as polysemy of words. If the algorithm determines the terms regardless of the semantic component of the words, considering homonymous words as one term, there is a los
Autor:
Oleg Olegovich Kondrashov, Aleksandra I. Kononova, Ilona I. Kapitanova, Oleg I. Lisov, Evgeny N. Petrov
Publikováno v:
2018 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus).
Current state of computer science enables one to automatize routine tasks of increasing number in various spheres. One of such spheres is bibliographical data which still implies routine manual processing that causes human mistakes and considerable t
Autor:
Vladimir F. Shangin, Andrey I. Kapitanov, Vladimir M. Troyanovskiy, Nikolay O. Krylikov, Ilona I. Kapitanova
Publikováno v:
2018 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus).
Terrorist and radical groups of people use instant messengers and accounts on social networks to publish propaganda messages. Blocking such accounts is one of the most effective methods of countering them. To do this, analysts need to read and proces
Autor:
Andrey I. Kapitanov, Elena L. Fedotova, Vladimir M. Troyanovskiy, Valentin V. Slyusar, Ilona I. Kapitanova
Publikováno v:
2017 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus).
In this paper we investigate the influence of outliers in the training set on the probabilistic classifier quality. By the example of naive Bayes classifier we show how the qualitative characteristics depend on the percentage of outliers' ratio. This