Taxonomic support for document classication in mobile device environment

This paper presents a novel document topic similarity measurement technique for mobile device environments. The goal is to help mobile device users to search for documents in a peer-to-peer environment which have similar topic to the documents on the users own device. The method is designed for slower processors, few memory and smaller data traffic between the devices to adapt to the environment of mobile devices like phones or PDA-s. The designed system employs a main server which helps the devices in the topic identification and representation. The keyword lists based topic comparison is enhanced with a taxonomic extension of the topic representations which allows comparison of documents with similar topics but few or no common keywords. The architecture, the employed algorithms and experimental results are presented in this paper.