The recommendations or guidelines about nutrition are available from a various distinct source on the internet. On the other hand, nutritional information needed by each person is different according to physical condition or personal preferences of each individual. This becomes a bit complicated because every information provider on the internet has a different understanding in giving foodstuff references to certain nutrients. In this study, an ontology in nutrition domain knowledge was used. The ontology represents explicit specification of pregnancy nutrition domain knowledge. The ontology constructed consists of three basic concepts that is Person, Maternal Condition and PrenNutriFood. To support the provision of nutritional recommendation, three definitions were added to ontology that is determining energy estimates per day, determining the percentage of daily value (DV) of food ingredients and determining the claims of nutrient content in foodstuff. In this study, we implemented the Semantic Web Rule Language to formalize those definitions. Inference from each rule is generated through Pellet as an inference engine. Ontology has been successfully managed with rules and finally produce new knowledge containing the recommendations. The results of inference indicate the expansion of knowledge in ontology