Examining feedback in interactive video retrieval

Author:

Albertson Dan1

Affiliation:

1. School of Library and Information Studies, University of Alabama, USA

Abstract

This paper presents a user-centred video retrieval, or video digital library, study that examines different types of feedback. Separate user interface features, including one that collects and one that presents feedback, shape the evaluation of this study. Research questions were posed to examine both types of feedback and to measure the influence of certain factors on each. The experimental factors examined here, in association with feedback, included user actions, topic characteristics, search duration, levels of user satisfaction and domain familiarity. Laboratory-style search experiments employing 28 users from the field of science education were conducted, which asked the users to attempt six pre-designed search topics. Actions of the users and topic durations were recorded, and users were given a post-search questionnaire about the topics and their experiences and impressions of the experimental system. Results showed that users regularly requested feedback about a clip, beyond a keyframe, while searching, not browsing, and that requests varied significantly across certain topic categories. Findings also indicated that users’ requests for feedback were more general interactions and not associated with a search topic’s representation of the domain. On the other hand, users rarely contributed feedback about the relative importance of visual vs semantic qualities of search results in order to reformulate queries, regardless of any other experimental factor.

Publisher

SAGE Publications

Subject

Library and Information Sciences,Information Systems

Cited by 3 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. A semantic-based video scene segmentation using a deep neural network;Journal of Information Science;2018-12-19

2. Weighted indexing of TV sports news videos;Multimedia Tools and Applications;2015-10-01

3. Visual information seeking;Journal of the Association for Information Science and Technology;2014-11-06

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3