Populating Ontologies with Data from OCRed Lists

A flexible, accurate, and efficient method of automatically extracting facts from lists in OCRed documents and inserting them into an ontology would help make those facts machine searchable, queryable, and linkable and expose their rich ontological interrelationships. To work well, such a process must be adaptable to variations in list format, tolerant of OCR errors, and careful in its selection of human guidance. We propose a wrapper-induction solution for information extraction that is specialized for lists in OCRed documents. In this approach, we induce a regular-expression grammar that can infer list structure and field labels in sequences of words in text. We decrease the cost and improve the accuracy of this induction process using semi-supervised machine learning and active learning, allowing induction of a wrapper from a single hand-labeled instance per field per list. To further reduce cost, we use the wrappers learned from the semi-supervised process to bootstrap an automatic (weakly supervised) wrapper induction process for additional lists in the same domain. In both induction scenarios, we automatically map labeled text to a rich variety of ontologically structured facts. We evaluate our implementation in terms of annotation cost and extraction quality for lists in historical documents.