TY - GEN
T1 - Personalized Extractive Summarization with Discourse Structure Constraints Towards Efficient and Coherent Dialog-Based News Delivery
AU - Takatsu, Hiroaki
AU - Ando, Ryota
AU - Honda, Hiroshi
AU - Matsuyama, Yoichi
AU - Kobayashi, Tetsunori
N1 - Funding Information:
Acknowledgements This work was supported by Japan Science and Technology Agency (JST) Program for Creating STart-ups from Advanced Research and Technology (START), Grant Number JPMJST1912 “Commercialization of Socially-Intelligent Conversational AI Media Service.”
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2022
Y1 - 2022
N2 - In this paper, we propose a method to generate a personalized summary that may be of interest to each user based on the discourse structure of documents in order to deliver a certain amount of coherent and interesting information within a limited time, primarily via a spoken dialog form. We initially constructed a news article corpus with annotations of the discourse structure, users’ profiles, and interests in sentences and topics. The proposed summarization model solves an integer linear programming problem with the discourse structure of each document and the total utterance time as constraints and extracts sentences that maximize the sum of the estimated degree of user’s interest. The degree of interest in a sentence is estimated based on the user’s profile obtained from a questionnaire and the word embeddings of BERT. Experiments confirm that the personalized summaries generated by the proposed method transmit information more efficiently than generic summaries generated based solely on the importance of sentences.
AB - In this paper, we propose a method to generate a personalized summary that may be of interest to each user based on the discourse structure of documents in order to deliver a certain amount of coherent and interesting information within a limited time, primarily via a spoken dialog form. We initially constructed a news article corpus with annotations of the discourse structure, users’ profiles, and interests in sentences and topics. The proposed summarization model solves an integer linear programming problem with the discourse structure of each document and the total utterance time as constraints and extracts sentences that maximize the sum of the estimated degree of user’s interest. The degree of interest in a sentence is estimated based on the user’s profile obtained from a questionnaire and the word embeddings of BERT. Experiments confirm that the personalized summaries generated by the proposed method transmit information more efficiently than generic summaries generated based solely on the importance of sentences.
UR - http://www.scopus.com/inward/record.url?scp=85142704913&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85142704913&partnerID=8YFLogxK
U2 - 10.1007/978-981-19-5538-9_4
DO - 10.1007/978-981-19-5538-9_4
M3 - Conference contribution
AN - SCOPUS:85142704913
SN - 9789811955372
T3 - Lecture Notes in Electrical Engineering
SP - 49
EP - 66
BT - Conversational AI for Natural Human-Centric Interaction - 12th International Workshop on Spoken Dialogue System Technology, IWSDS 2021
A2 - Stoyanchev, Svetlana
A2 - Ultes, Stefan
A2 - Li, Haizhou
PB - Springer Science and Business Media Deutschland GmbH
T2 - 12th International Workshop on Spoken Dialogue System Technology, IWSDS 2021
Y2 - 15 November 2021 through 17 November 2021
ER -