|Title||Discovery-based edit assistance for spreadsheets|
|Authors||Jacome Cunha, Joao Saraiva, & Joost Visser|
|Publication||IEEE Symposium on Visual Languages and Human-Centric Computing|
|Series||September, pages 233-237|
Spreadsheets can be viewed as a highly flexible end-users programming environment which enjoys wide-spread adoption. But spreadsheets lack many of the structured programming concepts of regular programming paradigms. In particular, the lack of data structures in spreadsheets may lead spreadsheet users to cause redundancy, loss, or corruption of data during edit actions.
In this paper, we demonstrate how implicit structural properties of spreadsheet data can be exploited to offer edit assistance to spreadsheet users. Our approach is based on the discovery of functional dependencies among data items which allow automatic reconstruction of a relational database schema.
From this schema, new formulas and visual objects are embedded into the spreadsheet to offer features for auto-completion, guarded deletion, and controlled insertion. Schema discovery and spreadsheet enhancement are carried out automatically in the background and do not disturb normal user experience.
Spreadsheets can be viewed as programming environments for non-professional programmers. We enhance spreadsheets with mechanisms to guide end-users to introduce correct data. Edit assistance is added to an existing spreadsheet based on a relational database schema obtained by data mining.