During travel planning, some travelers would first select the places which are famous or they are interested in, and then collect information for more travel spots to fill in the idle time. This scenario is must-visiting planning. However, the process is time-consuming and the sequence of the spots would affect travel experience. To support travelers in exploring spots efficiently without ruining the travel experience, we combine deep learning model - bidirectional LSTM and the concept of experience flow to provide candidates as interpolated spot to add between must-visiting spots. With this method, the travel plan can be enriched and experience can be enhanced.