Abstract

We present a new supervised framework that learns to estimate automatic Pyramid scores and uses them for optimization-based extractive multi-document summarization. For learning automatic Pyramid scores, we developed a method for automatic training data generation which is based on a genetic algorithm using automatic Pyramid as the fitness function. Our experimental evaluation shows that our new framework significantly outperforms strong baselines regarding automatic Pyramid, and that there is much room for improvement in comparison with the upper-bound for automatic Pyramid.

Highlights

  • We consider extractive text summarization, the task of condensing a textual source, e.g., a set of source documents in multi-document summarization (MDS), into a short summary text

  • Using the Text Analysis Conference (TAC)-2009 multi-document summarization dataset, we performed an upper-bound analysis for Automatic Pyramid (AP), and we evaluated the summaries extracted with our framework in an end-to-end evaluation using automatic evaluation metrics

  • We observed that the summaries extracted with our framework achieve significantly better AP scores than several strong baselines, but compared to the upper-bound for AP, there is still a large room for improvement

Read more

Summary

Introduction

We consider extractive text summarization, the task of condensing a textual source, e.g., a set of source documents in multi-document summarization (MDS), into a short summary text. Many state-of-the-art summarization systems cast extractive summarization as an optimization problem and maximize an objective function in order to create good, i.e., high-scoring summaries. To this end, optimization-based systems commonly use an objective function which encodes exactly those quality indicators which are measured by the particular evaluation metric being used. Humans annotate phrasal content units in a system summary and align them to the corresponding SCUs in the Pyramid set. The Pyramid score of a system summary is calculated as the sum of the SCU weights for all Pyramid set SCUs being aligned to annotated system summary phrases

Objectives
Methods
Results
Discussion
Conclusion
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.