Presentation 2005-02-25
Automatic generation of metadata for football games using announcer's commentary
Ichiro YAMADA, Masanori SANO, Hideki SUMIYOSHI, Masahiro SHIBATA, Nobuyuki YAGI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a method to generate metadata automatically for football games using announcer's commentary. Each commentary is classified using SVM based on the characteristics of the its structure. This clustering result expresses information of scene segmentation which shows a sequence of play event. Finally, events and the agent of each event are extracted from each scene. Experimental results using transcription written by human show the precision rate is 86.9% for the important event extraction, the recall rate is 65.4% in the event which can be extracted by human, and the accuracy for extracting agent of each event is 86.8%.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Metadata / SVM / Event Extraction / Scene Segmentation / Closed-caption Analysis
Paper # NLC2004-122,PRMU2004-204
Date of Issue

Conference Information
Committee NLC
Conference Date 2005/2/18(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Automatic generation of metadata for football games using announcer's commentary
Sub Title (in English)
Keyword(1) Metadata
Keyword(2) SVM
Keyword(3) Event Extraction
Keyword(4) Scene Segmentation
Keyword(5) Closed-caption Analysis
1st Author's Name Ichiro YAMADA
1st Author's Affiliation Science & Technical Research Laboratories, NHK()
2nd Author's Name Masanori SANO
2nd Author's Affiliation Science & Technical Research Laboratories, NHK
3rd Author's Name Hideki SUMIYOSHI
3rd Author's Affiliation Science & Technical Research Laboratories, NHK
4th Author's Name Masahiro SHIBATA
4th Author's Affiliation Engineering Administration Department, NHK
5th Author's Name Nobuyuki YAGI
5th Author's Affiliation Science & Technical Research Laboratories, NHK
Date 2005-02-25
Paper # NLC2004-122,PRMU2004-204
Volume (vol) vol.104
Number (no) 668
Page pp.pp.-
#Pages 6
Date of Issue