Exploration on predicting breast cancer stage with the aid of redesigned ANN incorporated with enhanced social spider optimisation technique

摘要

The core intention of this work is to predict the breast cancer stage as benignant or malignant from the given dataset with parameters such as instance clump thickness, uniformity of cell size, uniformity of cell shape, etc. Predicting the cancer stage helps to determine how to contain and eliminate breast cancer. One of the classification methods used is artificial neural network (ANN) which is trained with several training algorithms and the selected algorithm is Levenberg-Marquardt which performs better and gives minimum error value. To obtain the better prediction, the default structure of ANN is redesigned using optimisation techniques. To improve the structural design (hidden layer and neuron), diverse optimisation techniques are used, for example, Cuckoo search, Particle Swarm, Social Spider and Enhanced Social Spider Optimisation (ESSO). Our results show the ESSO is better and evaluate the metrics as Accuracy 97%, Sensitivity 98% and Specificity 95% compared with other techniques.