Ontology-Based Information Extraction from Handwritten Documents

Abstract

In this paper we introduce a new layer for the task of handwriting recognition. We add semantic information by means of ontologies. The task of our recognizer therefore is not only to recognize the ASCII transcription of the handwritten document, but also to identify the semantic concepts which appear in the text. This task is called ontology-based information extraction (OBIE), which has been applied to electronic documents recently. OBIE methods first segment the text into tokens, then identify their values and their corresponding instances of the ontology, and finally try to generate new facts based on the text. To the authors' knowledge, in this paper OBIE is proposed for the first time in handwriting literature. In our experiments we have evaluated the process up to the instantiation. We have found that using not only the top alternative, but also the k-best alternatives increases the performance of information extraction. Furthermore, the use of an ontology-based lexicon results in another performance increase.

From Handwriting Recognition to Ontologie-Based Information Extraction of Handwritten Notes

“…j ) is the belief value of the j-th instance in the path. In [6] experiments of ontology-based information extraction (OBIE) have shown that the proposed extension performs better than a simple approach which just feeds the output of the recognizer as an input to the OBIE system. Furthermore, using the ontology instances to alter the recognition lexicon increased the f-measure up to 69.67 % which is close to the performance on electronic text which is already available in ASCII.…”