Aggregation and Summarization of Thematically Similar Twitter Microblog Messages

Information is one of the most important resources in our modern lifestyle and society. Users on social network platforms, like Twitter, produce thousands of tweets every second in a continuous stream. However, not all written data are important for a follower, i.e., not necessary relevant information. That means, trawling through uncountable tweets is a time-consuming and depressing task, even if most of the messages are useless and do not contain news. This paper describes an approach for aggregation and summarization of short messages like tweets. Useless messages will be filtered out, whereas themost important information will be aggregated into a summarized output. Our experiments show the advantages of our promising approach, which can also be applied for similar problems.