An Author Gender Detection Method Using Whale Optimization Algorithm and Artificial Neural Network

Author gender detection (AGD) is a serious and crucial issue in Internet security applications, in particular in email, messenger, and social network communications. Detecting the gender of communication partner helps preventing massive fraud and abuses happening through social media such as email, blogs, forums. Text and writings of people on the Internet have valuable information that can be used to identify the gender of an author. Machine learning and meta-heuristic algorithms are valuable techniques to extract hidden patterns useful for detecting gender of a text. In this paper, an artificial neural network (ANN) is employed as a classifier to detect the gender of an email author and the whale optimization algorithm (WOA) is used to find optimal weights and biases for improving the accuracy of the ANN classification. Through this combination of ANN and WOA an accuracy of 98 & x0025;, precision of 97.16 & x0025;, and recall of 99.67 & x0025; were achieved, which indicates the superiority of the proposed method on Bayesian networks, regression, decision tree, support vector machine, and ANN examined.