Abstract

Tourism and hospitality have emerged as one of the pioneering sectors of sharing economy. However, homeowners who lack knowledge background are confused with pricing. Traditional hotel pricing and rental pricing methods may be not suitable for online accommodation rental. Therefore, CRISP-DM, the data analysis framework, is used to solve this problem. The price prediction model is established via the house data from the Airbnb.com. Finally, 33 determinants closely related to the price are found, and the most important 10 determinants are sorted. The study also finds several interesting rules: (1) the basic situation of housing is an important determinant, (2) online rental houses with more convenient transaction conditions have higher price, (3) providing more facilities and services can increase the price, (4) some determinants in traditional hotel pricing are not efficient in sharing houses. These findings can help the homeowners to understand customers and improve their own house and pricing.

Download

Share

COinS