PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Probabilistic Proactive Timeline Browser
Antti Ajanki and Samuel Kaski
In: 21st International Conference on Artificial Neural Networks (ICANN), 14-17 Jun 2011, Espoo, Finland.

Abstract

We have developed a browser suitable for finding events from timelines, in particular from life logs and other timelines containing a familiar narrative. The system infers the relevance of events based on the user’s browsing behavior and increases the visual saliency of relevant items along the timeline. As recognized images are strong memory cues, the user can quickly determine if the salient images are relevant and, if they are, it is quick and easy to select them by clicking since they are salient. Even if the inferred relevance was not correct, the timeline will help: The user may remember if the sought event was before or after a saliently shown event which limits the search space. A user study shows that the browser helps in locating relevant images quicker, and augmenting explicit click feedback with implicit mouse movement patterns further improves the performance.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Poster)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Information Retrieval & Textual Information Access
ID Code:8792
Deposited By:Antti Ajanki
Deposited On:21 February 2012