Shrimp and fish farming industry is one of the major productions that contributes importantly to the Vietnam economy. It provides hundreds of thousands of jobs and the hundred millions of U.S. dollars in revenues. However, this is also considered as a high-risk industry as shrimp and fish are very sensitive to the environmental changes. In addition, the development of diseases on these species is very fast. Any late treatment may cause serious problems to the species and consequently the reduction of revenues. Therefore, in this research, we propose an approach to diagnosing shrimp and fish diseases based on ontology. It helps the farmers making early predictions on shrimp and fish diseases from their morphological symptoms. The ontology is created manually based on the experts’ knowledge. The proposed system has been validated using a set of test cases. The validation result shows that the knowledge base and a set of rules have been implemented accurately.