2010 Presentation

Representing cell biology using a fixed set of relation names – a refined concept map approach



It is known that school level biology text is mostly of descriptive nature represented using the declarative form of knowledge. Biological knowledge can be characterized into concepts (terms, nouns) and relation names (linking words, verbs). There are hundreds of concepts spread across secondary school and undergraduate levels. Normally, several kinds of relation names are used in the representations. This approach leads to ambiguity in the propositions, and therefore lacks rigor. To overcome this problem, we have proposed a methodology to apply semantically accurate relation names explicitly and consistently to the concepts while mapping a domain.

In a study, we found that all relations found in the textbook could be represented using a specific set of relation names across all the levels (from school to undergraduate). It is generally accepted that conceptsare important for understanding of biology. However, that relation names are equally relevant is usually not so clear to teachers and learners. We state that concepts by themselves do not have any meaning; they attain meaning only when connected or linked with relation names to other concepts.

Therefore, our emphasis is characterizing biological knowledge by focussing on the relation names following the knowledge representation model. The principles applied are to use the relation names such as part-of, includes, surrounded by, located in, has function, example, has attributes, etc. This approach helps to resolve ambiguity in relations when using linking words like can be, have, are, maybe, etc. It also tends to be more rigorous and parsimonious in representation of scientific knowledge. In another study, it has been observed that domain experts not only use appropriate kinds of relation names while describing ideas within the domain but also use them consistently.

To determine what constitutes this fixed set of relation names in biology, we had undertaken to represent the biology textbook of Standards 8, 9 and 11 following the refined concept mapping approach. We know that, as the level of complexity increases from Standard 8, to 11, there would be an increase in the number of concepts, as more and more specific, complex concepts get introduced progressively. But even with this increasing complexity, we could determine a fixed set of relation names that could represent all the complexity. Therefore, we claim that if concepts are a measure of complexity, then relation names provide meaning to this complexity.

The presentation shall explore the refined concept mapping approach in finding the set of relation names with illustrations of concept maps based on textbook content in cell biology of Standards 8, 9 and 11.

Keywords: refined concept maps, concepts, relation names, knowledge representation, cell biology, scientific knowledge


Thesis advisor: Nagarjuna G.
Discussant: Chitra Natarajan