Data Processing Using Artificial Neural Networks

Abstract

The artificial neural network (ANN) is a machine learning (ML) methodology that evolved and developed from the scheme of imitating the human brain. Artificial intelligence (AI) pyramid illustrates the evolution of ML approach to ANN and leading to deep learning (DL). Nowadays, researchers are very much attracted to DL processes due to its ability to overcome the selectivity-invariance problem. In this chapter, ANN has been explained by discussing the network topology and development parameters (number of nodes, number of hidden layers, learning rules and activated function). The basic concept of node and neutron has been explained, with the help of diagrams, leading to the ANN model and its operation. All the topics have been discussed in such a scheme to give the reader the basic concept and clarity in a sequential way from ANN perceptron model to deep learning models and underlying types.

The accuracy of machine learning models using ultrasound images in prostate cancer diagnosis: A systematic review

“…This model can incorporate many variables and produce results in more complex situations. [45] In PCa diagnosis, ML can generate input data from various variables to classify whether the patient is suspected of having prostate cancer or not (Fig 3) .…”

“…[44] ML is a subdiscipline of AI where computer programs learn associations of predictive power from examples of data. [45] Several methods such as classification, regression, registration, and segmentation applied to analyse ultrasound images. However, neural networks algorithms have been shown to significantly improve performance when compared to other classifiers.…”

The Use of Machine-Learning Techniques in Material Constitutive Modelling for Metal Forming Processes

“…A set of biases b ∈ R m is added as the weight of a link, which in practice means incorporating a bias neuron that conventionally transmits a value of 1 to the output node [33]. In essence, in a FFNN, the n-dimensional input vector x is transformed into the outputs using the following recursive relations [33,75]:…”

“…There are several different types of activation functions, and their selection is an important part of the design process of a neural network. The most common activation functions [33,75] are shown in Figure A3. Worthy of note is the fact that all the activation functions shown are monotonic.…”

“…Unsupervised learning is concerned with finding patterns in unlabeled datasets containing many features. Supervised learning is the most widespread method used to train ANNs and uses labeled datasets, where each training sample is linked to a label [49,75]. The process is similar to a standard fitting procedure, accompanied by the selection of a learning algorithm that is used to fit target value [9,49].…”