Educational concept maps are generally less formal than other applications. This presents challenges for development of algorithms since there tends to be fewer features to compare. Similarity flooding is an algorithm that can compare available features by simultaneously matching structure and content between two concept maps.
An algorithm has been developed that recognizes patterns in the learner's responses presented in the map. The algorithm is designed to detect the learner's ability to connect similar concepts presented in the learned material and the learner's use of relationships that indicate acquisition of the desired content.
The disambiguation process is an approach that requires assigning meaning to every word in context. The algorithm assigns the meaning based on the assumed meaning in terms of semantics, and contextual meaning within the language from which the word is derived. The goal of the approach is to overcome the limited information provided by the structure of the concept map.