Semi-automatic extraction of cross-table data from a set of spreadsheets
Alaaeddin Swidan (TU Delft - Software Engineering)
Felienne Hermans (TU Delft - Software Engineering)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Spreadsheets are widely used in companies. End-users often value the high degree of flexibility and freedom spreadsheets provide. However, these features lead to the development of a variety of data forms inside spreadsheets. A cross-table is one of these forms of data. A cross-table is defined as a rectangular form of data, which expresses the relations between a set of objects and a set of attributes. Cross-tables are common in spreadsheets: our exploratory analysis found that more than 3.42% of spreadsheets in an industrial open dataset include at least one cross-table. However, current software tools provide no support to analyze data in cross-tables. To address this, we presents a semi-automatic approach to extract cross-table data from a set of spreadsheets, and transform them to a relational table form. We evaluate our approach in a case study, on a set of 333 spreadsheets with 2,801 worksheets. The results show that the approach is successful in extracting over 92% of the data inside the targeted cross-tables. Further, we interview two users of the spreadsheets working in the company; they confirmed the approach is beneficial and provides correct results.