Novel rules for extracting the entities of entity relationship models

Extracting entities from natural language text to design conceptual models of the entity relationships is not trivial and novice designers students can find it especially difficult. Researchers have suggested linguistic rules/guidelines for extracting text. Unfortunately, while these guidelines are often correct they can, also, be invalid. There no rule that true at all times. This paper suggests novel rules based on machine learning classifiers, RIPPER, PART decision trees. Performance comparison was made between rules. The results shows there a dramatic improvement when were used.