Abstract

In the context of rural revitalization, rural land transfer is an important way to promote economic development in poor areas, and it is urgent to carry out an in-depth analysis related to rural land transfer in poor mountainous areas. This paper proposes to construct an IAFSA-ELM prediction model of rural land transfer based on ELM algorithm and IAFSA algorithm, and set evaluation indexes based on sample data. Based on the comparative analysis of the subjective reasons for willingness to transfer land, the highest percentage of the reasons for willingness to transfer is "V3 (high cost, hard work, low income)" (68.4%), while the highest percentage of the reasons for unwillingness to transfer is "M1 (need to secure food rations) The highest percentage of the reasons for not willing to transfer is "M1 (need to guarantee food rations)", which is 50.7%. This is a preliminary judgment that farmers' willingness to transfer land is closely related to their livelihood needs. To study the accuracy comparison results of IAFSA-ELM land transfer optimization prediction model, the accuracy rate of IAFSA-ELM algorithm for land transfer optimization prediction model stays in the range of 84%-92%, and in terms of accuracy above, IAFSA-ELM algorithm performs better compared to the two algorithms alone. This study is beneficial to the optimization of rural land transfer, thus promoting rural land transfer in poor mountainous areas and improving land use efficiency.