{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}


brezeale-kdd2007-LearningVideoPreferencesFromVideoContent -...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Learning Video Preferences from Video Content Darin Brezeale Department of Computer Science and Engineering The University of Texas at Arlington Box 19015, Arlington, TX 76019, USA [email protected] Diane J. Cook School of Electrical Engineering and Computer Science Washington State University Pullman, WA 99164-2752, USA [email protected] ABSTRACT Viewers of video now have more choices than ever. As the number of choices increases, the task of searching through these choices to locate video of interest is becoming more dif- ficult. Current methods for learning a viewer’s preferences in order to automate the search process rely either on video having content descriptions or on having been rated by other viewers identified as being similar. However, much video ex- ists that does not meet these requirements. To address this need, we use hidden Markov models to learn the preferences of a viewer by combining visual features and closed cap- tions. Results are provided from some initial experiments using this approach. Categories and Subject Descriptors H.3.1 [ Information Storage and Retrieval ]: Content Analysis and Indexing; I.2.6 [ Artificial Intelligence ]: Learn- ing Keywords video preferences, user modeling, closed captions 1. INTRODUCTION People today have access to more video than at any time in history. Sources of video include television broadcasts, movie theaters, movie rentals, video databases, and the In- ternet. While many video choices come from the entertain- ment domain, other types of video are becoming more com- mon, such as educational lectures at universities and confer- ences [34]. As the number of video choices increases, the task of searching for video of interest is becoming more difficult. One approach that viewers take is to search for video within specific genre. In the case of entertainment video, the genre of the video is provided when the video is released. How- ever, there is much video that is unclassified. This has led to research in automatically classifying video by genre. While Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. MDM/KDD’07 , August 12, 2007, San Jose, California, USA. Copyright 2007 ACM X-XXXXX-XXX-X ... $ 5.00. knowing the genre of video is helpful, the large amounts of video choices within many genre still make finding video of interest a time-consuming process. In addition, this prob- lem is even greater for people who enjoy video from a variety of genre, which seems likely for most people. For these rea- sons, systems have been developed that can learn a partic- ular person’s preferences and make recommendations given these preferences.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}