A study of the classification problem in context of information theory is presented in the paper.... more A study of the classification problem in context of information theory is presented in the paper. Current research in that field is focused on optimisation and bayesian approach. Although that gives satisfying results, they require a vast amount of data and computations to train on. Authors propose a new concept, named Informational Neurobayesian Approach (INA), which allows to solve the same problems, but requires significantly less training data as well as computational power. Experiments were conducted to compare its performance with the traditional one's and the results showed that capacity of the INA is quite promising.
The article presents a method that improves the quality of classification of objects described by... more The article presents a method that improves the quality of classification of objects described by a combination of known and unknown features. The method is based on modernized Informational Neurobayesian Approach with consideration of unknown features. The proposed method was developed and trained on 1500 text queries of Promobot users in Russian to classify them into 20 categories (classes). As a result, the use of the method allowed to completely solve the problem of misclassification for queries with combining known and unknown features of the model. The theoretical substantiation of the method is presented by the formulated and proved theorem On the Model with Limited Knowledge. It states, that in conditions of limited data, an equal number of equally unknown features of an object cannot have different significance for the classification problem.
In the following paper the authors present a GAN-type model and the most important stages of its ... more In the following paper the authors present a GAN-type model and the most important stages of its development for the task of emotion recognition in text. In particular, we propose an approach for generating a synthetic dataset of all possible emotions combinations based on manually labelled incomplete data. 1 Dataset vectorization Initially, it is necessary to form a dataset for training the model. The dataset should contain data about seven basic emotions for each certain piece of text-a paragraph or a sentence. The authors couldn't find a publicly available dataset with this or similar data. Therefore, it was
Bulletin of the Moscow State Regional University, 2017
Аннотация. 26 апреля 2017 г. на факультете политологии МГУ имени М.В. Ломоносова состоялся круглы... more Аннотация. 26 апреля 2017 г. на факультете политологии МГУ имени М.В. Ломоносова состоялся круглый стол «Глобальная и региональная безопасность в условиях меняющегося миропорядка». В процессе дискуссии обсуждался вопрос о необходимости переосмысления категории «безопасность» с учётом новых вызовов и угроз действующему мировому порядку. Региональная безопасность была рассмотрена в глобальном измерении, в контексте её влияния на мирополитические процессы. Интересы участников были сосредоточены на анализе политики безопасности России, на тенденциях евразийской интеграции, а также на деятельности межгосударственных объединений. Результатом обмена опытом стало формирование актуального представления о глобальной и региональной безопасности. Модератором круглого стола выступил профессор П.А. Цыганков.
A study of the classification problem in context of information theory is presented in the paper.... more A study of the classification problem in context of information theory is presented in the paper. Current research in that field is focused on optimisation and bayesian approach. Although that gives satisfying results, they require a vast amount of data and computations to train on. Authors propose a new concept, named Informational Neurobayesian Approach (INA), which allows to solve the same problems, but requires significantly less training data as well as computational power. Experiments were conducted to compare its performance with the traditional one's and the results showed that capacity of the INA is quite promising.
The article presents a method that improves the quality of classification of objects described by... more The article presents a method that improves the quality of classification of objects described by a combination of known and unknown features. The method is based on modernized Informational Neurobayesian Approach with consideration of unknown features. The proposed method was developed and trained on 1500 text queries of Promobot users in Russian to classify them into 20 categories (classes). As a result, the use of the method allowed to completely solve the problem of misclassification for queries with combining known and unknown features of the model. The theoretical substantiation of the method is presented by the formulated and proved theorem On the Model with Limited Knowledge. It states, that in conditions of limited data, an equal number of equally unknown features of an object cannot have different significance for the classification problem.
In the following paper the authors present a GAN-type model and the most important stages of its ... more In the following paper the authors present a GAN-type model and the most important stages of its development for the task of emotion recognition in text. In particular, we propose an approach for generating a synthetic dataset of all possible emotions combinations based on manually labelled incomplete data. 1 Dataset vectorization Initially, it is necessary to form a dataset for training the model. The dataset should contain data about seven basic emotions for each certain piece of text-a paragraph or a sentence. The authors couldn't find a publicly available dataset with this or similar data. Therefore, it was
Bulletin of the Moscow State Regional University, 2017
Аннотация. 26 апреля 2017 г. на факультете политологии МГУ имени М.В. Ломоносова состоялся круглы... more Аннотация. 26 апреля 2017 г. на факультете политологии МГУ имени М.В. Ломоносова состоялся круглый стол «Глобальная и региональная безопасность в условиях меняющегося миропорядка». В процессе дискуссии обсуждался вопрос о необходимости переосмысления категории «безопасность» с учётом новых вызовов и угроз действующему мировому порядку. Региональная безопасность была рассмотрена в глобальном измерении, в контексте её влияния на мирополитические процессы. Интересы участников были сосредоточены на анализе политики безопасности России, на тенденциях евразийской интеграции, а также на деятельности межгосударственных объединений. Результатом обмена опытом стало формирование актуального представления о глобальной и региональной безопасности. Модератором круглого стола выступил профессор П.А. Цыганков.
Uploads
Papers by Ivan Bolokhov