Cross-kingdom regulation of human genes by plant miRNAs has always been a topic of discussion and attention. Several reports have been published on the presence of dietary plant miRNAs in the human digestive tract, raising great hopes for their discovery as potential therapeutics. Recent studies have demonstrated that plant miRNAs can enter the bloodstream and plasma of mammals through the digestive tract, subsequently reaching tissues and cells. miRNAs are a type of single-stranded non-coding RNAs with a length of approximately 19–24 nucleotides, which are known as potent regulators of gene expression. These small molecules can regulate mRNAs by cleavage or inhibition of translation. Therefore, discovering the targets of miRNAs is of great importance in biological research. However, due to their short length and limited sequence complementarity with target genes, developing algorithms for accurate prediction of miRNA targets has always been challenging. In this study, the Kronecker method and the mintRULS algorithm, the regularized least squares objective function, were used to predict the targets of plant miRNAs in the human body. Thus, by the effective features that are used in mintRULS algorithm and applying them in the plant miRNA and human mRNA sequence datasets in a similarity-based manner, we were able to achieve a very promising level of model performance in the experimental dataset from 0.839 to 0.842 in identifying human target sites and classifying them into three classes of strong, medium, and weak, which indicates its high generalizability For the purpose of prognosis and treatment of various diseases.