A multi-level system for sequential update summarization

Chunyun Zhang, Zhanyu Ma, Jiayue Zhang, Weiran Xu, Jun Guo

Abstract: When an emergency occurs, such as the outbreak of natural disaster, the news about the incident showed a trend of blowout. Mostly, these news are reproduced by people and spread out with duplicate, unimportant or wrong information. Hence, it is necessary and vital to provide individuals with timely and important information of these incidents during their development. In this paper, we present a new multi-level system which can broadcasts with useful, new, and timely sentence-length updates about a developing event. The new system proposed a novel method, which incorporates techniques from topic-level and sentence-level summarization. To evaluate the performance of the proposed system, we applied it to the task of sequential update summarization of Temporal Summarization (TS) track at Text Retrieval Conference (TREC). Experimental results showed that our proposed method have a good performance.