Supporting Searching on Small Screen Devices Using Summarisation

Abstract

In recent years, small screen devices have seen widespread increase in their acceptance and use. Combining mobility with their increased technological advances many such devices can now be considered mobile information terminals. However, user interactions with small screen devices remain a challenge due to the inherent limited display capabilities. These challenges are particularly evident for tasks, such as information seeking. In this paper we assess the effectiveness of using hierarchical-query biased summaries as a means of supporting the results of an information search conducted on a small screen device, a PDA. We present the results of an experiment focused on measuring users' perception of relevance of displayed documents, in the form of automatically generated summaries of increasing length, in response to a simulated submitted query. The aim is to study experimentally how users' perception of relevance varies depending on the length of summary, in relation to the characteristics of the PDA interface on which the content is presented. Experimental results suggest that hierarchical query-biased summaries are useful and assist users in making relevance judgments.

New Directions in Cognitive Information Retrieval

“…The advantages of using previously defined relevance assessments are that we can identify for what types of relevant documents the system is performing poorly, that we can perform some error checking as was done in the interactive track of TREC, and we can define alternative evaluation measures, e.g. Sweeney and Crestani (2003) used decision correctness: the proportion of relevant and non-relevant documents correctly identified, a measure that is only possible when using ground truth test collections.…”

‘Show me more’: Incremental length summarisation using novelty detection

“…By adopting a strategy to detect novelty in the generation of summaries, we aim to reduce the amount of redundant information contained in subsequent summaries. In previous studies, we have investigated the use of hierarchical query-biased summarisation using summaries of increasing length on a mobile phone (Sweeney, Crestani, & Tombros, 2002) and a PDA (Sweeney & Crestani, 2003). In those studies our assumption of 'show me more' was simply as providing a summary of increasing length.…”

“…Sweeney et al (2002;Sweeney & Crestani, 2003), investigate the use of automatically-generated hierarchical query-biased text summaries of newspaper articles presented to on WAP mobile phones, and PDAs. They describe hierarchical text summaries as having a root, or top level summary, which corresponds to the minimum level of information; each hierachy, or summary level, is then intended to provide more information.…”

Physical Document Adaptation to User’s Context and User’s Profile

“…There have been some attempts to calculate the optimal size of the document's summary. Sweeny (Sweeney 2003) has mentioned some static percentage to generate an adapted document. For example he suggests that a summary of 3% is the most optimal value for the mobile phones (without considering the size of the document!).…”