A pattern based tokenization model for XML parsing on mobile devices

Abstract

This paper presents a theoretical tokenization model for XML parsing on resource constrained mobile devices. The model is based on the identification of sequentially repeating patterns within the structure of an XML document. As soon as it identifies a repeating structure, it relieves the parser from the computationally intensive conventional tokenization process, and focuses on extracting text node based values for further processing by the calling application. Our experiments demonstrate that the proposed tokenization model considerably relieves the processing bottlenecks encountered in conventional XML parsers.