Microsummarization of Online Reviews: An Experimental Study

Authors

  • Rebecca Mason Google, Inc.
  • Benjamin Gaska University of Arizona
  • Benjamin Van Durme Johns Hopkins University
  • Pallavi Choudhury Microsoft Research
  • Ted Hart Microsoft Research
  • Bill Dolan Microsoft Research
  • Kristina Toutanova Microsoft Research
  • Margaret Mitchell Microsoft Research

DOI:

https://doi.org/10.1609/aaai.v30i1.10396

Keywords:

summarization, entity recognition, sentiment analysis, opinion mining, recommendations

Abstract

Mobile and location-based social media applications provide platforms for users to share brief opinions about products, venues, and services. These quickly typed opinions, or microreviews, are a valuable source of current sentiment on a wide variety of subjects. However, there is currently little research on how to mine this information to present it back to users in easily consumable way. In this paper, we introduce the task of microsummarization, which combines sentiment analysis, summarization, and entity recognition in order to surface key content to users. We explore unsupervised and supervised methods for this task, and find we can reliably extract relevant entities and the sentiment targeted towards them using crowdsourced labels as supervision. In an end-to-end evaluation, we find our best-performing system is vastly preferred by judges over a traditional extractive summarization approach. This work motivates an entirely new approach to summarization, incorporating both sentiment analysis and item extraction for modernized, at-a-glance presentation of public opinion.

Downloads

Published

2016-03-05

How to Cite

Mason, R., Gaska, B., Van Durme, B., Choudhury, P., Hart, T., Dolan, B., Toutanova, K., & Mitchell, M. (2016). Microsummarization of Online Reviews: An Experimental Study. Proceedings of the AAAI Conference on Artificial Intelligence, 30(1). https://doi.org/10.1609/aaai.v30i1.10396

Issue

Section

Technical Papers: NLP and Text Mining