文章基本信息

标题：On Novel summarization and Time line Generation for Evolutionary Tweet Streams
本地全文：下载
作者：Revathi ; P. Sujitha ; V. Janardhan Babu 等
期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2017
卷号：8
期号：3
页码：430-433
出版社：TechScience Publications
摘要：As increasing popularity of social siteslikeTweeter, Facebook and Instagram etc. We will get lotof tweets and „N‟ no. of short messages being sharedatunpredictable rate which is very high. As this data islargeenough it will become critical to understand andanalyzetherefore redundancy and noisy data must beremoved. To overcome these drawbacks of existing systemwe proposesumblr framework, in comparison with otherregular approaches of summarization which depends onstatic data and small datasets where sumblr is dynamicand works on large data set. Firstly we have proposedtweet cluster vector algorithm for maintaining statisticaldata and compact cluster information to maintaindynamically in memory during stream processing, storeand organize cluster snapshots of different moments.Generation of online and historical summaries witharbitrary time durations, we propose TCV ranksummarization algorithm. We have proposed anevaluation method which generates timeline,categorization based on topic evaluation.
关键词：Tweet stream; continuous summarization;summary; timeline