Title
A Study of Video-based Concordancer on Scene Classification
Abstract
Video-based concordancer can be used to provide a scenario that engage context information of studying novel words. Usually, a fixed number of contextual sentences are retrieved accompanying the keywords. However, there may be a lack of complete context for learners to comprehend the keywords in the videos. Few studies have discussed about how the videos are presented to assist learners to use the keywords appropriately, and lead learners to find relevant knowledge effectively. In this paper, a keyword-in-scene video concordance (KWIS), which recognizes the scenes in the videos and provides the scene-based clips, is proposed. Each video clip is tagged actual scene type information. Learners are able to query the KWIS system with keywords, phrases, or natural language sentences, and watch relevant scenario clips to understand where the conversation can be carried on. A pilot study was conducted to evaluate the proposed system. The result shows that there is a positive effect on students' comprehension of English phrases while using the system.
Year
DOI
Venue
2011
10.1109/ICALT.2011.30
ICALT
Keywords
Field
DocType
scene-based clips,video signal processing,video-based concordancer,corpus,scene classification,contextual sentence retrieval,context information,video concordancer,information retrieval,video clip,vocabulary,film-based learning,actual scene type information,relevant scenario clip,english phrases,english phrase,full-text retrieval,complete context,image classification,proposed system,kwis system,natural language sentences,scene detection,natural language processing,text analysis,relevant knowledge,keyword-in-scene video concordance,computer aided instruction,natural language,color,visualization,semantics,films,mathematical model
Conversation,Concordancer,Computer science,Visualization,Natural language,Artificial intelligence,Natural language processing,Contextual image classification,Multimedia,Vocabulary,Comprehension,Semantics
Conference
ISSN
ISBN
Citations 
2161-3761 E-ISBN : 978-0-7695-4346-8
978-0-7695-4346-8
0
PageRank 
References 
Authors
0.34
7
4
Name
Order
Citations
PageRank
Tz-Yung Fang100.34
Jie-Chi Yang235043.91
Yu-Chieh Wu324723.16
Chi-cheng Tsai4191.91