Table motif extraction device, classifier learning device, table type classification device, method, and program

摘要

PROBLEM TO BE SOLVED: To extract features capable of improving extraction accuracy of table types from tabular data.SOLUTION: An n-th order neighboring cell extraction part 22 extracts a cell set including a cell and an n-th order cell neighboring the cell with respect to respective cells included in a piece of input tabular data; and a cell identifier applying part 24 applies a cell identifier to a cell included in the extracted cell set based on the structure and the content of the cell with respect to the respective cells included in the tabular data. A table motif output section 26 outputs a motif feature which represents a piece of information of the cell set based on the cell identifier applied to each cell in the cell set with respect to each of the extracted cell sets on each of the cells.SELECTED DRAWING: Figure 1