To generate an abstractive summary for a large set of semantic associations, we select and verbalize a diverse set of semantic association patterns. As part of our MEAST toolkit, our tool called MEAST-SUMM is open-source under the Apache License. Note that our tool depends on stanford-corenlp-3.7.0.jar and stanford-corenlp-3.7.0-models.jar, which are not included in our release. [Download]
We also provide the 80 queries used in our experiments. Each file contains one query; in the file, each line provides the URI of one entity. [Download]
How to cite:
Yu Gu, Yue Liang, Gong Cheng, Daxin Liu, Ruidi Wei, Yuzhong Qu.
Diversified and Verbalized Result Summarization for Semantic Association Search.
In Proceedings of the 19th International Conference on Web Information Systems Engineering (WISE'18).
An extended technical report is also available. [Download]
Contact Gong Cheng with any questions or comments.