The occurrence and development of several human physiological processes are significantly influenced by the competing endogenous RNA (ceRNA) network. The aim of the present study was to construct a microRNA (miRNA)-mRNA network associated with exosomes in ovarian cancer (OV), and experimental validation of key target genes. By exploring the Gene Expression Omnibus (GEO) database, we analyzed the RNAs from 226 samples to identify differentially expressed miRNAs (DEMs) and genes (DEGs) that showed differential expression as OV progressed. Subsequently, we conducted Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses on the DEGs. Furthermore, we constructed a miRNA-mRNA network that pertains to exosomes in OV using DEMs and DEGs. Moreover, we validated the expression levels of mRNAs in the miRNA-mRNA network using Gene Expression Profiling Interactive Analysis (GEPIA2). Ultimately, luciferase reporter assay was used to identify the potential target relationship between FYVE and coiled-coil domain containing 1 (FYCO1) and miRNAs. Our analysis screened a total of 14 DEMs and 101 DEGs, and the DEGs were mainly enriched in DNA replication or repair, amino acid biosynthesis and carbon metabolism. Furthermore, a miRNA-mRNA network was constructed including 3 miRNAs (hsa-miR-17-5p, hsa-miR-20b-5p and hsa-miR-20a-5p) and 2 mRNAs, FYCO1 and purine rich element binding protein A (PURA). Finally, the 2 mRNAs in this miRNA-mRNA network were verified by GEPIA2 using The Cancer Genome Atlas (TCGA) database. Among them, only FYCO1 showed significant different expression of mRNA in OV and normal tissue, while the prognosis of FYCO1 in OV remains controversial due to different database. Interestingly, FYCO1 was identified as the target of hsa-miR-17-5p. By constructing a novel network of miRNA-mRNA, we can gain new understanding of the molecular mechanisms that drive exosomes in OV. Targeting FYCO1, which originates from exosomes, may hold promise as a diagnostic marker for OV.