Powered by OpenAIRE graph
Found an issue? Give us feedback

Исследование эффективности решения задачи классификации гибридными сетями Кохонена

Исследование эффективности решения задачи классификации гибридными сетями Кохонена

Abstract

В статье приводятся результаты решения задачи классификации стандартной гибридной сетью Кохонена и предложенными авторами модификациями данной модели: составной, распределённой и нечёткой гибридными сетями Кохонена. Исследования проводились на модельных задачах репозитория UCI. Приведены структуры предложенных нейросетевых моделей, описаны методики их обучения. Исследована эффективность решения задачи классификации гибридными сетями Кохонена на основе анализа двух критериев: погрешности классификации и суммарного числа нейронов в сети, при котором достигнута минимальная погрешность классификации.

The results of solution of classification using standard hybrid kohonen neural network and modifications of this model created by authors composite, distributed and fuzzy hybrid kohonen neural network were obtained in this paper. Researches were performed on dataset models from UCI repository. Neural networks’ structures and teaching algoritms were described. The efficiency of neural networks based on two criterions: classification error and neurons number in network when minimum classification error was archived was researched.

Keywords

ГИБРИДНАЯ СЕТЬ КОХОНЕНА,СОСТАВНАЯ ГИБРИДНАЯ СЕТЬ КОХОНЕНА,РАСПРЕДЕЛЁННАЯ ГИБРИДНАЯ СЕТЬ КОХОНЕНА,НЕЧЁТКАЯ СЕТЬ КОХОНЕНА,АЛГОРИТМ WTA,АЛГОРИТМ WTM,АЛГОРИТМ ОБРАТНОГО РАСПРОСТРАНЕНИЯ ОШИБКИ,АЛГОРИТМ C-MEANS,HYBRID KOHONEN NEURAL NETWORK,COMPOSITE HYBRID KOHONEN NEURAL NETWORK,DISTRIBUTED HYBRID KOHONEN NEURAL NETWORK,FUZZY KOHONEN NEURAL NETWORK,WTA ALGORITHM,WTM ALGORITHM,BACKPROPAGATION,C-MEANS ALGORITHM

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
gold