Work place: Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia, 86400, Parit Raja, Batu Pahat, Johor, Malaysia
Research Interests: Neural Networks, Swarm Intelligence, Data Mining, Data Structures and Algorithms
Rozaida Ghazali is currently a Professor at the Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia (UTHM). She graduated with a Ph.D. degree in Higher Order Neural Networks from the School of Computing and Mathematical Sciences at Liverpool John Moores University, United Kingdom in 2007. Earlier, in 2003 she completed her M.Sc. degree in Computer Science from Universiti Teknologi Malaysia (UTM). She received her B.Sc. (Hons) degree in Computer Science from Universiti Sains Malaysia (USM) in 1997. In 2001, Rozaida joined the academic staff in UTHM. Her research area includes neural networks, swarm intelligence, optimization, data mining, and time series prediction. She has successfully supervised a number of PhD and master students and published more than 100 articles in various international journals and conference proceedings. She acts as a reviewer for various journals and conferences, and as an editor in a few Springer conference proceedings. She has also served as a conference chair, and as a technical committee for numerous international conferences.
DOI: https://doi.org/10.5815/ijisa.2020.06.02, Pub. Date: 8 Dec. 2020
Financial time-series prediction has been long and the most challenging issues in financial market analysis. The deep neural networks is one of the excellent data mining approach has received great attention by researchers in several areas of time-series prediction since last 10 years. “Convolutional neural network (CNN) and recurrent neural network (RNN) models have become the mainstream methods for financial predictions. In this paper, we proposed to combine architectures, which exploit the advantages of CNN and RNN simultaneously, for the prediction of trading signals. Our model is essentially presented to financial time series predicting signals through a CNN layer, and directly fed into a gated recurrent unit (GRU) layer to capture long-term signals dependencies. GRU model perform better in sequential learning tasks and solve the vanishing gradients and exploding issue in standard RNNs. We evaluate our model on three datasets for stock indexes of the Hang Seng Indexes (HSI), the Deutscher Aktienindex (DAX) and the S&P 500 Index range 2008 to 2016, and associate the GRU-CNN based approaches with the existing deep learning models. Experimental results present that the proposed GRU-CNN model obtained the best prediction accuracy 56.2% on HIS dataset, 56.1% on DAX dataset and 56.3% on S&P500 dataset respectively.[...] Read more.
DOI: https://doi.org/10.5815/ijisa.2020.05.05, Pub. Date: 8 Oct. 2020
The standard method to train the Higher Order Neural Networks (HONN) is the well-known Backpropagation (BP) algorithm. Yet, the current BP algorithm has several limitations including easily stuck into local minima, particularly when dealing with highly non-linear problems and utilise computationally intensive training algorithms. The current BP algorithm is also relying heavily on the initial weight values and other parameters picked. Therefore, in an attempt to overcome the BP drawbacks, we investigate a method called Modified Cuckoo Search-Markov chain Monté Carlo for optimising the weights in HONN and boost the learning process. This method, which lies in the Swarm Intelligence area, is notably successful in optimisation task. We compared the performance with several HONN-based network models and standard Multilayer Perceptron on four (4) time series datasets: Temperature, Ozone, Gold Close Price and Bitcoin Closing Price from various repositories. Simulation results indicate that this swarm-based algorithm outperformed or at least at par with the network models with current BP algorithm in terms of lower error rate.[...] Read more.
DOI: https://doi.org/10.5815/ijitcs.2017.05.04, Pub. Date: 8 May 2017
In this paper, new statistical features based approach (SFBA) for hourly energy consumption prediction using Multi-Layer Perceptron is presented. The model consists of four stages: data retrieval, data pre-processing, feature extraction and prediction. In the data retrieval stage, historical hourly consumed energy data has been retrieved from the database. During data pre-processing, filters have been applied to make the data more suitable for further processing. In the feature extraction stage, mean, variance, skewness, and kurtosis are extracted. Finally, Multi-Layer Perceptron has been used for prediction. For experimentation with Multi-Layer Perceptron with different training algorithms, a final model of the network was designed in which the scaled conjugate gradient (trainscg) was used as a network training function, tangent sigmoid (Tansig) as a hidden layer transfer function and linear function as an output layer transfer function. For hourly energy consumption prediction, a total of six weeks data of ten residential buildings has been used. To evaluate the performance of the proposed approach, Mean Absolute Error (MAE), Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), evaluation measurements were applied.[...] Read more.
Subscribe to receive issue release notifications and newsletters from MECS Press journals