Forecasting the Unemployment Rate by Neural Networks Using Search Engine Query Data

Web information, regarded as a useful repository including abundant data and knowledge, attracts much attention from researchers and practitioners, and has been used to analyze and forecast economic and social hotspots in recent years. In this paper, a novel neural network based forecasting method is proposed for the unemployment rate prediction using search engine query data. The empirical results show that the proposed method outperforms other forecasting methods, which have been used for the unemployment rate prediction. These findings imply that web information, especially web search behavior, can improve the efficiency and effectiveness of the unemployment rate prediction.