Real Estate Confidence Index Based on Real Estate News
Methods of text classification are applied to research on the construction of RECI by applying the Naïve Bayes algorithm to evaluate data and to classify the extent to which this measure describes confidence in the REM.
Abstract
ABSTRACT A real estate confidence index (RECI) is used to evaluate real estate industry development, and it has become an effective and powerful measure in China’s real estate market (REM). RECI research based on big data is the new trend in finance and economics. In this article, we apply some methods of text classification to research on the construction of RECI. First, the Naïve Bayes algorithm is used to evaluate data and to classify the extent to which this measure describes confidence in the REM. Second, experiments on different perspectives are performed to probe the relationship between variables and the accuracy of the classifier. Third, we use the classifier to predict the weekly news. Ultimately, construction of the RECI based on financial and economic news is achieved by applying the classifier to the time and existence of major financial and economic news.