Conceptual-model-based data extraction from multiple-record Web pages
Data & Knowledge Engineering
Multilingual ontologies for cross-language information extraction and semantic search
ER'11 Proceedings of the 30th international conference on Conceptual modeling
Hi-index | 0.00 |
Most computer software available today -- although capable of processing text, numbers and other symbols -- cannot process meaning or nuances as we people do. One such case, which is the focus of this work, is the processing of number-rich documents, such as receipts to fulfill business-specific constraints. This task currently is labor intensive in the sense that we manually analyze the receipts and file the information in the required categories of business expenses reports. The premises of this paper is that it should be possible to overcome many of the limitations present in manual, number-rich information filing applications, by further understanding the lexical and semantic relationships through extraction ontologies. This paper introduces one such approach and describes preliminary experimental results that support this hypothesis.