Peter Jörgensen

Image querying by image professionals

This article reports the analysis of two samples of search logs from a commercial image provider over a 1-month period. The study analyzes image searches and queries, user query modification strategies, and user browsing and downloading of results. Unique term searches are less frequent than earlier research has shown; descriptive and thematic queries are more common. Boolean searching, although heavily employed, appears to be ineffective and leads to query modifications. Although there was a large amount of query modification (61.7% of queries across the two samples), the tactics overall do not appear to be carefully thought out and seem to be largely experimental. Given the willingness to modify queries but the inability to do so in an effective way, more support for query modification may be beneficial.

Incorporating context in text analysis by interactive activation with competition artificial neural networks

Many segments of modern society, including marketing, politics, government, activism and public safety, desire the ability to find relationships, thus meaning, in public discourse. This can be accomplished by analyzing communication documents according to their content. The increasing use of the Internet for public dialog has made Internet communication a potentially rich source of data in this regard. This study explores the use of an interactive activation with competition (IAC) artificial neural network (ANN) to find relationships in email texts. A modified fully recurrent IAC network was used to process 69 email messages from two threads in the Open Library/Information Science Education Forum using two variations of the self-organizing phase of network formation. These variations were: (1) with and (2) without a linear decay function applied between sentences to the external activation of nodes. The use of the linear decay function, which could be considered a method for including context, produced three positive effects: the entire network was more differentiated from keywords; the keywords were more highly associated with each other, and; roughly half the number of noise strings were highly associated with keywords.

A flexible image browsing interface tester: Design, implementation and preliminary results in a face recognition test

This paper describes a software tool for exploring factors that affect image browsing. The tool, which can present images in storyboard or slideshow formats, can easily be configured to change the size of the images, the display rate, and other parameters of interest. Results of a preliminary experiment involving 26 subjects using the slideshow mode support earlier findings and are reported.