China is facing an aging society, and more and more disabled elders need Long-Term Care(LTC) from care-givers. On the other hand, care-givers can obtain some income from providing their care services. Two-sided matching decision-making method is a very useful tool to deal with this social problem. In this paper, we propose a novel multi-period two-sided matching decision-making method for probabilistic linguistic information using D-S evidence theory. Firstly, we use D-S evidence theory to aggregate the information for different attributes. By D-S evidence theory, we can reduce the information loss in the process of aggregating the information in different attributes. Then, we use a programming to compute the time weighting vector, which can solve the dynamic decision making problem. Furthermore, we propose a novel method to obtain the value perception function based on prospect theory, by which the bounded rationality of decision maker can be considered. Finally, we apply our method to a real case in rural village in Zhenjiang City, Jiangsu Province to illustrate the effectiveness of our method.INDEX TERMS Multi-period two-sided matching, probabilistic linguistic term set, D-S evidence theory, prospect theory, disabled elder.