Column-Type Prediction for Web Tables Powered by Knowledge Base and Text

Web tables are essential for applications such as data analysis. However, web often incomplete and short of some critical information, which makes it challenging to understand the table content. Automatically predicting column types without metadata is significant dealing with various from Internet. This paper proposes a CNN-Text method deal this task, fuses CNN prediction voting processes. We present augmentation synthetic generation approaches improve CNN’s performance use extracted text get better predictions. The experimental result shows that outperforms baseline methods, demonstrating well qualified type prediction.