A Proposal to Extend Concept Mapping to Concept Lattices for Representing Biology Meena Kharatmal Homi Bhabha Centre for Science Education (TIFR), Mumbai, India meena@hbcse.tif.res.in Summer School on “Methodology of task design – How to construct exercises for learning” Technische Universität Dresden, Dresden, Germany 22 – 27 September 2014. Abstract. Textbook biology knowledge is being represented as triples using concept maps. These triples can be extended to create concept lattices by representing objects and attributes in relation. By focusing on the nature of semantic relations, concept neighbourhood lattices can be generated for dependencies, associations. By representing the changes in the attributes of objects in time, concept lattices of dynamic propositions can be generated. As the knowledge base for the study is textbook biology, this research on using concept lattices in school, college biology education can be further developed with focusing on teaching learning, cognitive assessment. Keywords. Concept mapping, concept lattices, formal concept analysis, assessment, knowledge structure, biology Formal Concept Lattice for Knowledge Discovery Formal Concept Analysis (FCA) is a theory of data analysis which identifies conceptual structures among data sets. The FCA is considered as a method to structure, analyze and visualize data for its implications and dependencies. Knowledge Discovery (KD), is the nontrivial extraction of implicit, previously unknown, and potentially useful information from data. A concept is a unit of thought comprising of extension ie all objects and intension ie all attributes of the objects. The objects and attributes are related in hierarchical, subconcept- superconcept relations, the implication between attributes, and incidence relation of object having attributes. A group that combines the set of objects (G) with the attributes (M) in a relation (I) is called a formal context defined as a triple of (G, M, I) (Ganter & Wille, 1997). A formal context is generally represented as a binary incidence table in which the crosses represent the binary relation between the object set and the attribute set. Figure 1 shows a table of a formal context with a concept lattice indicating the relations of objects and their attributes. A proposal for Representing Textbook Biology Using FCA The texbook biology knowledge has been represented as simple propositions in the form of triples (Kharatmal & Nagarjuna, 2010). These triples have been extracted by and represented using the semantic relations referring the relations ontology, of the open biomedical domain (Smith, et al., 2005). Some of these relations belong to part- whole, class-inclusion, spatial inclusion, function, attribution relations, cause-effect, etc. 1. Creating Concept Lattices Based on Semantic Relations in Biology: Textbook biology knowledge is mostly descriptive. Representing the same graphically would enhance its understanding at school, college levels. One of the ways is using concept maps – a two dimensional graphical representation using node-link-node form to create triples. Figure 2 shows a concept map showing concepts connected with semantic relations. Concept lattices on lexical relations have been researched on the WordNet (Priss, 2010). Based on concept neighbourhoods, 1 Figure 1: A illustration of formal context and concept lattice. (Source: Uta Priss, FCA Website, http://www.upriss.org.uk/fca/fcaintro. html ) Figure 2: A Concept Map showing concepts connected with linking phrases (semantic relations).