Auto-completion for Data Cells in Relational Tables
This work addresses data management and analysis challenges for users dealing with incomplete relational tables, though it is incremental as it builds on existing auto-completion methods with specific enhancements.
The paper tackles the problem of auto-completing data cells in relational tables by introducing the CellAutoComplete framework, which handles multiple conflicting values, provides supporting evidence, and manages empty cells, resulting in a 40% improvement over the best baseline.
We address the task of auto-completing data cells in relational tables. Such tables describe entities (in rows) with their attributes (in columns). We present the CellAutoComplete framework to tackle several novel aspects of this problem, including: (i) enabling a cell to have multiple, possibly conflicting values, (ii) supplementing the predicted values with supporting evidence, (iii) combining evidence from multiple sources, and (iv) handling the case where a cell should be left empty. Our framework makes use of a large table corpus and a knowledge base as data sources, and consists of preprocessing, candidate value finding, and value ranking components. Using a purpose-built test collection, we show that our approach is 40\% more effective than the best baseline.