A new denoising approach based on mode decomposition applied to the stock market time series: 2LE-CEEMDAN.

Time series, including noise, non-linearity, and non-stationary properties, are frequently used in prediction problems. Due to these inherent characteristics of time series data, forecasting based on this data type is a highly challenging problem. In many studies within the literature, high-frequency components are commonly excluded from time series data. However, these high-frequency components can contain valuable information, and their removal may adversely impact the prediction performance of models. In this study, a novel method called Two-Level Entropy Ratio-Based Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (2LE-CEEMDAN) is proposed for the first time to effectively denoise time series data. Financial time series with high noise levels are utilized to validate the effectiveness of the proposed method. The 2LE-CEEMDAN-LSTM-SVR model is introduced to predict the next day's closing value of stock market indices within the scope of financial time series. This model comprises two main components: denoising and forecasting. In the denoising section, the proposed 2LE-CEEMDAN method eliminates noise in financial time series, resulting in denoised intrinsic mode functions (IMFs). In the forecasting part, the next-day value of the indices is estimated by training on the denoised IMFs obtained. Two different artificial intelligence methods, Long Short-Term Memory (LSTM) and Support Vector Regression (SVR), are utilized during the training process. The IMF, characterized by more linear characteristics than the denoised IMFs, is trained using the SVR, while the others are trained using the LSTM method. The final prediction result of the 2LE-CEEMDAN-LSTM-SVR model is obtained by integrating the prediction results of each IMF. Experimental results demonstrate that the proposed 2LE-CEEMDAN denoising method positively influences the model's prediction performance, and the 2LE-CEEMDAN-LSTM-SVR model outperforms other prediction models in the existing literature.

The use of NARX neural network for modeling of adsorption of zinc ions using activated almond shell as a potential biosorbent.

In this study, nonlinear autoregressive model processes with exogenous input (NARX) are applied for the prediction of percentage adsorption efficiency for the removal of zinc ions from wastewater by activated almond shell. The effect of operational parameters such as pH, dosage, particle size and initial metal ions concentration are studied to optimize the conditions for maximum removal of zinc ions. The model is first developed using a two layer NARX network. A comparison between the model results and experimental data showed that the NARX model is able to predict the removal of zinc ions from wastewater. The outcomes of suggested NARX modeling were then compared to batch experimental studies. The results show that activated almond shell is an efficient sorbent and NARX network, which is easy to implement and is able to model the batch experimental system.

A fast and adaptive automated disease diagnosis method with an innovative neural network model.

Automatic disease diagnosis systems have been used for many years. While these systems are constructed, the data used needs to be classified appropriately. For this purpose, a variety of methods have been proposed in the literature so far. As distinct from the ones in the literature, in this study, a general-purpose, fast and adaptive disease diagnosis system is developed. This newly proposed method is based on Learning Vector Quantization (LVQ) artificial neural networks which are powerful classification algorithms. In this study, the classification ability of LVQ networks is developed by embedding a reinforcement mechanism into the LVQ network in order to increase the success rate of the disease diagnosis method and reduce the decision time. The parameters of the reinforcement learning mechanism are updated in an adaptive way in the network. Thus, the loss of time due to incorrect selection of the parameters and decrement in the success rate are avoided. After the development process mentioned, the newly proposed classification technique is named "Adaptive LVQ with Reinforcement Mechanism (ALVQ-RM)". The method proposed handles data with missing values. To prove that this method did not offer a special solution for a particular disease, because of its adaptive structure, it is used both for diagnosis of breast cancer, and for diagnosis of thyroid disorders, and a correct diagnosis rate after replacing missing values using median method over 99.5% is acquired in average for both diseases. In addition, the success rate of determination of the parameters of the proposed "LVQ with Reinforcement Mechanism (LVQ-RM)" classifier, and how this determination affected the required number of iterations for acquiring that success rate are discussed with comparison to the other studies.

Comparison of the results of response surface methodology and artificial neural network for the biosorption of lead using black cumin.

In this study, Response Surface Methodology (RSM) and Artificial Neural Network (ANN) were employed to develop an approach for the evaluation of heavy metal biosorption process. A batch sorption process was performed using Nigella sativa seeds (black cumin), a novel and natural biosorbent, to remove lead ions from aqueous solutions. The effects of process variables which are pH, biosorbent mass, and temperature, on the sorbed amount of lead were investigated through two-levels, three-factors central composite design (CCD). Same design was also utilized to obtain a training set for ANN. The results of two methodologies were compared for their predictive capabilities in terms of the coefficient of determination-R(2) and root mean square error-RMSE based on the validation data set. The results showed that the ANN model is much more accurate in prediction as compared to CCD.