The speed of learning in neural network environment is considered as the most effective parameter spatially in large data sets. This paper tries to minimize the time required for the neural network to fully understand and learn about the data by standardize input data. The paper showed that the Z-Score standardization of input data significantly decreased the number of epoochs required for the network to learn. This paper also proved that the binary dataset is a serious limitation for the convergence of neural network, so the standardization is a must in such case where the 0’s inputs simply neglect the connections in the neural network. The data set used in this paper are features extracted from gel electrophoresis images and that open the door for using artificial intelligence in such areas.

No abstract

Artificial intelligence has many branches of image processing-based applications in terms of classification and identification, error back-propagation neural network is a great match for such applications as long as linear vector quantization (LVQ) and pattern recognition is another great match for recognition of digital images based on their features. The dataset used in this paper are gel electrophoresis images where 6 features had been extracted from the images and used as input to a neural network for learning and then checked for recognition purposed and the system managed to recognize all the 6 images. Six features had been used: average, standard deviation, smoothness, skewness, uniformity, and entropy. A tiny error rate where allowed in the recognition program to cover the variation of the dataset and the test data (gel-electrophoresis images). The proposed system had successfully managed to identify all of the learned data in both LVQ and error-back-propagation. Error-back-propagation proved itself as a great tool in terms of learning time compared with LVQ, which was very slow in terms of learning time and recognition.