A hybrid swarm intelligence approach for blog success prediction

Abstract

Successful blogs receive high ratings and generate marketing value. What factors contribute to the success of a blog and how to predict its success level are questions worth discussing. A hybrid swam intelligence approach is proposed in this study to predict blog success level. First, this study develops a research model of blog success with six factors from content, technology, and social views of point, which include currentness, design, reliability, security, interaction, and connectivity. A questionnaire is designed based on the blog success model. Two hundred ten valid samples are collected from Internet users with experience in using or creating blogs. A hybrid approach combining particle swarm optimization (PSO) and self-organizing map (SOM) is proposed to predict blog success level. The results of 10-fold validation are examined to compare the hybrid PSO-SOM approach with the results from three classifiers: C5.0, classification and regression trees (CARTs), and support vector machine (SVM). For blog success prediction, the results indicate the PSO-SOM approach demonstrates higher accuracy among these methods.