具有终端用户服务的在线系统:从电视广播字幕中挖掘新概念

Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining Pub Date : 2013-08-11 DOI:10.1145/2487575.2487715

Mika Rautiainen, Jouni Sarvanko, A. Heikkinen, M. Ylianttila, V. Kostakos

{"title":"具有终端用户服务的在线系统:从电视广播字幕中挖掘新概念","authors":"Mika Rautiainen, Jouni Sarvanko, A. Heikkinen, M. Ylianttila, V. Kostakos","doi":"10.1145/2487575.2487715","DOIUrl":null,"url":null,"abstract":"Better tools for content-based access of video are needed to improve access to time-continuous video data. Particularly information about linear TV broadcast programs has been available in a form limited to program guides that provide short manually described overviews of the program content. Recent development in digitalization of TV broadcasting and emergence of web-based services for catch-up and on-demand viewing bring out new possibilities to access data. In this paper we introduce our data mining system and accompanying services for summarizing Finnish DVB broadcast streams from seven national channels. We describe how data mining of novelty concepts can be extracted from DVB subtitles to augment web-based \"Catch-Up TV Guide\" and \"Novelty Cloud\" TV services. Furthermore, our system allows accessing media fragments as Picture Quotes via generated word lists and provides content-based recommendations to find new programs that have content similar to the user selected programs. Our index consists of over 180 000 programs that are used to recommend relevant programs. The service has been under development and available online since 2010. It has registered over 5000 user sessions.","PeriodicalId":20472,"journal":{"name":"Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining","volume":"102 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2013-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"An online system with end-user services: mining novelty concepts from tv broadcast subtitles\",\"authors\":\"Mika Rautiainen, Jouni Sarvanko, A. Heikkinen, M. Ylianttila, V. Kostakos\",\"doi\":\"10.1145/2487575.2487715\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Better tools for content-based access of video are needed to improve access to time-continuous video data. Particularly information about linear TV broadcast programs has been available in a form limited to program guides that provide short manually described overviews of the program content. Recent development in digitalization of TV broadcasting and emergence of web-based services for catch-up and on-demand viewing bring out new possibilities to access data. In this paper we introduce our data mining system and accompanying services for summarizing Finnish DVB broadcast streams from seven national channels. We describe how data mining of novelty concepts can be extracted from DVB subtitles to augment web-based \\\"Catch-Up TV Guide\\\" and \\\"Novelty Cloud\\\" TV services. Furthermore, our system allows accessing media fragments as Picture Quotes via generated word lists and provides content-based recommendations to find new programs that have content similar to the user selected programs. Our index consists of over 180 000 programs that are used to recommend relevant programs. The service has been under development and available online since 2010. It has registered over 5000 user sessions.\",\"PeriodicalId\":20472,\"journal\":{\"name\":\"Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining\",\"volume\":\"102 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-08-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2487575.2487715\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2487575.2487715","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

为了改进对时间连续视频数据的访问，需要更好的基于内容的视频访问工具。特别是关于线性电视广播节目的信息，已经以一种仅限于节目指南的形式提供了简短的手动描述的节目内容概述。电视广播数字化的最新发展和基于网络的追播和点播服务的出现，为获取数据提供了新的可能性。本文介绍了我们的数据挖掘系统和配套服务，用于总结芬兰七个国家频道的DVB广播流。我们描述了如何从DVB字幕中提取新颖性概念的数据挖掘，以增强基于web的“追赶电视指南”和“新颖性云”电视服务。此外，我们的系统允许通过生成的单词列表访问图片引用的媒体片段，并提供基于内容的推荐，以查找与用户选择的节目内容相似的新节目。我们的索引包含超过18万个节目，用于推荐相关节目。这项服务自2010年以来一直在开发中，并可在线使用。它已经注册了5000多个用户会话。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

An online system with end-user services: mining novelty concepts from tv broadcast subtitles

Better tools for content-based access of video are needed to improve access to time-continuous video data. Particularly information about linear TV broadcast programs has been available in a form limited to program guides that provide short manually described overviews of the program content. Recent development in digitalization of TV broadcasting and emergence of web-based services for catch-up and on-demand viewing bring out new possibilities to access data. In this paper we introduce our data mining system and accompanying services for summarizing Finnish DVB broadcast streams from seven national channels. We describe how data mining of novelty concepts can be extracted from DVB subtitles to augment web-based "Catch-Up TV Guide" and "Novelty Cloud" TV services. Furthermore, our system allows accessing media fragments as Picture Quotes via generated word lists and provides content-based recommendations to find new programs that have content similar to the user selected programs. Our index consists of over 180 000 programs that are used to recommend relevant programs. The service has been under development and available online since 2010. It has registered over 5000 user sessions.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining

自引率

0.00%

发文量