Health informatics data can be conserving by high level integrity of storage. When improving complexity and potential causality of such data, semi-automated approach with intelligo-ontology process used to improve the precision. Machine learning and NLP technique used for information-absence recovery on each non-taxonomic data. Annotation can be m de by each NLP progress and it will distinguished by ontology with its linked relationships. The algorithm uses EMR data to identify the absence of fundamental relationships between symptoms and disorders in contextual knowledge and suggests believable relation-ships that can rectify missing relationships using semantics of the domain concepts. The co-occurrence-based method will have a better recall at the cost of precision; it considers all co-occurring disorders as candidates to explain the symptom. In summary, the algorithm enables making effective use of domain experts for building high quality knowledge bases. The proposed method solves as much better than the NLP engine to annotate the entities and associate negation and temporal information with the entity.