\r\ntechnique in artificial neural network and has been used as a tool

\r\nfor solving the time series problems, such as decreasing training

\r\ntime, maximizing the ability to fall into local minima, and optimizing

\r\nsensitivity of the initial weights and bias. This paper proposes an

\r\nimprovement of a BP technique which is called IM-COH algorithm

\r\n(IM-COH). By combining IM-COH algorithm with cuckoo search

\r\nalgorithm (CS), the result is cuckoo search improved control output

\r\nhidden layer algorithm (CS-IM-COH). This new algorithm has a

\r\nbetter ability in optimizing sensitivity of the initial weights and bias

\r\nthan the original BP algorithm. In this research, the algorithm of

\r\nCS-IM-COH is compared with the original BP, the IM-COH, and the

\r\noriginal BP with CS (CS-BP). Furthermore, the selected benchmarks,

\r\nfour time series samples, are shown in this research for illustration.

\r\nThe research shows that the CS-IM-COH algorithm give the best

\r\nforecasting results compared with the selected samples.","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 147, 2019"}