A Case-Based Reasoning Approach for Unknown Class Invoice Processing

This paper introduces an invoice analysis approach using case-based reasoning (CBR). CBR is used to analyze and interpret new invoices thanks to the previous processing experiences. Each new document is segmented into structures and interpreted thanks to a structure database. Interpreting a new document's structures relies on graph edit distance as well as on string edit distance. This paper focuses on document structure extraction as well as on document interpretation via its structures interpretation. The proposed system reaches an extraction and interpretation rate of 76.33%.