Identifying Cyberspace Users’ Tendency in Blog Writing Using Machine Learning Algorithms

A blog is a form of direct interactive communication technology, which allows users to interact and communicate with each other through posting comments and sharing links as well. A blog is a platform where a writer or group of writers gives their opinion on a specific topic. Many issues and topics that are in a certain country being censored and controlled by the government from being presented through the mass media. Nevertheless, blogs have the space to provide a wide platform for exchanging ideas and opinions on various issues. There is a specific proportion between blog features and bloggers’ tendency to social, political, and cultural patterns of different countries and nations that create trends among the bloggers in these countries. In this paper, we use an existing data set from previous research, which has 100 records of data, and manipulate the data by applying three machine learning algorithms for implementing classification and regression tasks. The algorithms are Decision Tree (c4.5), Linear Regression (LR), and Decision Forest (DF) with a 10-fold cross-validation method for training and testing. The results showed that C4.5 achieves the best overall results of 81% accuracy, 83% precision, and 91% recall, compared with the other two algorithms.