Paper: Efficient Online Summarization of Microblogging Streams

ACL ID E14-4046
Title Efficient Online Summarization of Microblogging Streams
Venue Annual Meeting of The European Chapter of The Association of Computational Linguistics
Session Main Conference
Year 2014

The large amounts of data generated on microblogging services are making sum- marization challenging. Previous research has mostly focused on working in batches or with filtered streams. Input data has to be saved and analyzed several times, in or- der to detect underlying events and then summarize them. We improve the effi- ciency of this process by designing an on- line abstractive algorithm. Processing is done in a single pass, removing the need to save any input data and improving the run- ning time. An online approach is also able to generate the summaries in real time, us- ing the latest information. The algorithm we propose uses a word graph, along with optimization techniques such as decaying windows and pruning. It outperforms the baseline in terms of summary quality, as well as t...