Corpus ID: 201637006

Compressive Multi-document Summarization with Sense-level Concepts ∗

@inproceedings{Shen2019CompressiveMS,
  title={Compressive Multi-document Summarization with Sense-level Concepts ∗},
  author={X. Shen and Wai Lam and Xunying Liu and Piji Li},
  year={2019}
}
  • X. Shen, Wai Lam, +1 author Piji Li
  • Published 2019
  • Most existing document summarization methods make use of information appeared in input documents only. They operate on the word level, and do not make use of the sense-level concepts which usually need external semantic knowledge. In this paper, we investigate a compressive MultiDocument Summarization (MDS) model which integrates the sense-level concepts of words and entities. With sense disambiguation of the text, we propose a novel way of salience estimation and redundancy reduction that… CONTINUE READING

    Tables from this paper

    References

    SHOWING 1-10 OF 28 REFERENCES
    Statistics-Based Summarization - Step One: Sentence Compression
    • Kevin Knight, D. Marcu
    • Computer Science
    • AAAI/IAAI
    • 2000
    • 441
    • PDF
    Reader-Aware Multi-Document Summarization: An Enhanced Model and The First Dataset
    • 9
    • PDF
    Text summarization using Wikipedia
    • 49
    Do Multi-Sense Embeddings Improve Natural Language Understanding?
    • 187
    • PDF
    Entity Linking meets Word Sense Disambiguation: a Unified Approach
    • 680
    • PDF
    Word sense disambiguation: A survey
    • 1,710
    • Highly Influential
    • PDF