TEAM 3: Semantic Cleaning of the SPOI Database

Project leaders: Raul Palma rpalma(at) and Karel Panek

Team members: Otakar Čerba (University of West Bohemia)

The goal will be to use semantic technologies to identify duplicities in SPOI in case of data coming from different resources and with different attributes. Data will be compared and checked against other information sources to improve general quality of SPOI database. The proposed approach will also integrate advanced parsing technology, knowledge base with extensive model of selected natural language(s), and selected classification methods. Along with consistent fuzzy duplicates detection, the technology provides a robust foundation for dynamic categorisation and verification, i. e. against more diverse data sources and across languages.