Semi-automatic knowledge transformation of semantic network ontologies into Frames structures
Sajid Ullah Khan, Muhammad Khan, Muhammad Asif Nauman · 2016
Knowledge is modeled in various knowledge representation formats like semantic network, decision table, decision tree, etc. Such representations are used to express knowledge in natural languages. There is a major difference between programming languages and Natural languages. They both are used for communication purpose. The former is used for communication between human and machine and the latter is used for human communication. Semantic network is one of the knowledge representation technique used for communication between knowledge engineer and user. It lacks logical completeness and exactness. Due to the uncertainty of information about nodes and links in semantic networks problems arise in inferring and queering of knowledge. There is an object oriented layout of knowledge representation known as Frames structures that can be modified with slot filling capability and procedural attachments. In this research paper, a solution is proposed to extract knowledge from semantic networks ontologies in a semi-automatic way of knowledge transformation into an inferring suitable knowledge source so-called frame structure. Our proposed technique readout a standard Owl XML file expressing semantic network ontology designed in Protege, and transforms into an equivalent frame based knowledge source. The resulted knowledge source is stored in Frames repository providing better inferring capability. Various case examples have been adopted in order to validate the proposed technique by verifying the number and names of the nodes/frames, the number and type of attributes and methods for each node/frame and the type of relationship between any of the two nodes/frames were identical in the input and the corresponding output.