Alzheimer’s disease (AD), a neurodegenerative disease, is closely related to the genetic and environmental factors, about 70% of which are caused by the genetic factors, but its pathogenesis is still unclear. Along with the advent of high-throughput gene sequencing, the processing of genetic data using machine learning (ML) has become a hot spot. In this paper, the applications of ML in AD are mainly reviewed, including the diagnosis and prognosis of AD based on genetic data, the analysis of genetic variation of AD, the analysis of gene expression profile of AD, and the further development of ML for AD. Firstly, during the diagnosis and prognosis of AD, the genetic data combining with other modalities, such as imaging data, clinical data and histological data, would be greatly improved the accuracy of ML methods. It is valuable for the early diagnosis of AD, and effectively delays the progression of AD. Secondly, the application of ML in the analysis of genetic variation of AD, single nucleotide polymorphisms (SNPs) of new genes were dug out, and the pathogenic mechanism of AD was further explored. Thirdly, the analysis of gene expression profile of AD mainly focuses on the discovery of the pathways of genes which could provide the possibility of gene targets for AD therapy. In the future, the multi-level model of ML might be developed for high-quality, diverse and large data, and provide scientific strategies for exploring the pathogenesis of AD.