User Driven Example

The task of creating a lexical knowledgebase has been defined in our work as extracting appropriate semantic phenomena from what are essentially print dictionaries stored in a desktop publishing format. The aim of this work is to achieve automatic identification of structural elements in the dictionaryy’s stream of text that are isomorphic to semantic fields of interest. A fully automatic solution appears infeasible to this problem. A semi-automatic approach that exploits training from a knowledgeable user to identify structures is shown to be fruitful. Once learnt from the user the structures are then applied automatically to other text streams in the same document or to other documents. On completion of training the data records are written into a lexical knowledgebase in an XML mark up format. Experimental work demonstrates the effectiveness of the system by a user training the system on a minimal set of entries in a dictionary. The usefulness of the learned rules is demonstrated with a 4 fold increase in processing throughput after a training of only 178 records in 138 minutes. Key features of the software system, Ferret III, are ease of use, fast performance and robustness of the system.