Abstract: With the rapid growth of China's green consumer groups, green consumption has shown great environmental value while stimulating consumption upgrades. The impact of energy-saving information exposure on green consumer behavior has also attracted scholars' attention. This paper combines natural language processing(NLP) technology to quantify the energy-saving information exposure index, and adopts propensity score matching(PSM) and quantile regression(QR) to conduct an empirical research of 5,889 household appliance consumption data and 604,759 comments text data from June 2017 to October 2017 on a large home appliance sales platform in China. It is found that the exposure of energy-saving information has a significant effect on increasing product sales. With the increase in sales volume, the effect of energy-saving information exposure on energy-saving appliance consumption is “U”-type change; at the same time, the overall energy-saving information of China's current e-commerce platform The exposure is not high, and the marginal effect of China's energy-saving information exposure on energy-saving appliance sales has shown a monotonous upward trend; in addition, this paper further validates the existence of “platform effect”.

Key words: Energy-saving information exposure, NLP, PSM, Green consumer behavior