Lung cancer is a leading global cause of cancer-related death, and lung adenocarcinoma (LUAD) accounts for ~ 50% of lung cancer. Here, we screened for novel and specific biomarkers of LUAD by searching for differentially expressed mRNAs (DEmRNAs) and microRNAs (DEmiRNAs) in LUAD patient expression data within The Cancer Genome Atlas (TCGA). The identified optimal diagnostic miRNA biomarkers were used to establish classification models (including support vector machine, decision tree, and random forest) to distinguish between LUAD and adjacent tissues. We then predicted the targets of identified optimal diagnostic miRNA biomarkers, functionally annotated these target genes, and performed receiver operating characteristic curve analysis of the respective DEmiRNA biomarkers, their target DEmRNAs, and combinations of DEmiRNA biomarkers. We validated the expression of selected DEmiRNA biomarkers by quantitative real-time PCR (qRT-PCR). In all, we identified a total of 13 DEmiRNAs, 2301 DEmRNAs and 232 DEmiRNA-target DEmRNA pairs between LUAD and adjacent tissues and selected nine DEmiRNAs (, , , , , , , , and ) as optimal LUAD-specific biomarkers with great diagnostic value. The predicted targets of these nine DEmiRNAs were significantly enriched in transcriptional misregulation in cancer and central carbon metabolism. Our qRT-PCR results were generally consistent with our integrated analysis. In summary, our study identified nine DEmiRNAs that may serve as potential diagnostic biomarkers of LUAD. Functional annotation of their target DEmRNAs may provide information on their roles in LUAD.