Темир йўл транспорти: долзарб масалалар ва инновациялар 3-tom (2024) · 22-32-betler
NEURAL NETWORK MODELS FOR PREDICTING TRAFFIC FLOWS: NEURAL NETWORK MODELS FOR PREDICTING TRAFFIC FLOWS
Жалелов, Р.М.
Annotaciya
The increasing number of vehicles leads to different levels of traffic congestion. Although many researchers have proposed many solutions to solve the problem of traffic congestion, there is still no ideal solution to solve it. Because the chaotic complexity of the transport network, population mobility and many other complex factors have a negative impact on the effective management of traffic flows. This paper investigates the issue of forecasting and managing traffic flows, which uses methods based on deep neural networks, in particular, Elman's recurrent neural network (RNN), long short-term memory (LSTM), Gated Recurrent Unit (GRU). The models are used in experiments, and their mean square error (MSE) is estimated by the evaluation criterion. According to the results of computational experiments, the GRU model is recognized as the most effective model.
транспортный поток, нейронная сеть, временной ряд, Элман, LSTM, GRU, нормализация, функция активации, глубокое обучение, набор тестов.traffic flow, neural network, time series, Elman, LSTM, GRU, normalization, activation function, deep learning, test set.
Metadata derekkózi: jurnal OAI-PMH arxivi · Sindex tolıq mátindi saqlamaydı, derekkózge silteme beredi.