Extraction Algorithms for Hierarchical Header Structures from Spreadsheets.
Keisuke Goto, Yuiko Ohta, Hiroya Inakoshi, Nobuhiro Yugami · 2016
Spreadsheets are widely used to manage statistical data, and a massive amount of valuable spreadsheets are now public as open data. Spreadsheet headers may contain hierarchical structures, and values in spreadsheets are associated with not only headers at the same column or row but also those corresponding to their ancestors. When integrating spreadsheets with other datasets, it is necessary to extract hierarchical header structures and associate this information with values. In this paper, we propose three algorithms to extract hierarchical structures from spreadsheets based on the relation of values. Since our algorithms are focused only on values, they can be applied to heterogeneous spreadsheets such that the structures of row or column headers are not explicitly in