TabSum: a flexible and dynamic table summarization approach

Abstract

Many diverse small devices, such as smart phones and personal digital assistants, are being deployed to access the Internet. Small devices typically have limited display and processing capabilities. It is thus difficult to effectively present a large table of information on these different devices such that the table can be easily browsed by the users. In this paper, we present the design and prototype implementation of TabSum, a flexible and dynamic table summarization approach to reducing tables into smaller but still meaningful representations for various devices. TabSum supports easy browsing of a large table by various devices and provides personalization by taking into account both device capabilities and user preferences. A multi-source multi-layered specification methodology is proposed to describe summarization rules preferred by various sources. Upon requests, a set of table reduction rules are dynamically applied to the rows and columns of a table to reduce its size. The approach is flexible and dynamic. It summarizes both numeric and non-numeric data types.