|
In the eyes of Chinese people, real estate is synonymous with family. Buying a house is a milestone in life. A house protects a family from wind and rain. It can cover independent individuals and have private space. It can be financial management. It is often heard in friends' occasions at dinners and family gatherings, and many people have gained a feeling in buying houses, so housing prices are often the topic of daily topics.
This paper mainly uses the housing price data of Linkou District, New Taipei City. Due to the economic development of this area in recent years, housing price has always been a topic of concern in Taipei City and New Taipei City. The data is obtained from the real-price registration data provided by the Ministry of the Interior's real estate transaction information supply system and the economic-related data on the statistical information website of the Republic of China. The time axis starts and ends from January 2020 to December 2021. A total of 2 years, because the data during this period is close to the latest state and relatively complete. This paper refers to the previous column of housing price prediction models, and use machine learning to construct models with the same variables. In this paper, random forest regression, support vector regression, K-nearest neighbor algorithm, decision tree regression and the stacking method of the integration of four algorithms, a total of five ways to build a housing price prediction model. The results show that the error rate of the random forest regression method and the stacking method is the same, but compared with the reference paper, the model error has not been improved. It is speculated that the reason is that the data is not cut and classified like the reference paper.
|