Identifying Gender of Internet Users Based on Access History

The use of activities and internet access differs between men and women. On average, men spend more time on the Internet a day. Men also have some of the same online activities as women. However, there are specific differences such as men’s tendency to access features such as breaking news, football, or games and men’s products. On the contrary, women are more interested in shopping, e-commerce, chatting and participating in social networking sites and blogs. The study aims to identify and predict gender of internet users based on their access history. With SVM method, the correct classification rate is the highest compared to the other two models Accuracy = 87.67%, in addition, the Precision, Recall, and F-Score parameters also give outstanding rates. This result allows us to believe in the ability of the SVM machine learning model to effectively handle the classification and gender identification problem with large-dimensional data.