Automatic expressive opinion sentence generation for enjoyable conversational systems

Yoichi Matsuyama, Akihiro Saito, Shinya Fujie, Tetsunori Kobayashi

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

In terms of functional conversations, Grice's Maxim of Quantity suggests that responses should contain no more information than was explicitly asked for. However, in our daily conversations, more informative response skills are usually employed in order to hold enjoyable conversations with interlocutors. These responses are usually produced as forms of one's additional opinions, which usually contain their original viewpoints as well as novel means of expression, rather than simple and common responses characteristic of the general public. In this paper, we propose automatic expressive opinion sentence generation mechanisms for enjoyable conversational systems. The generated opinions are extracted from a large number of reviews on the web, and ranked in terms of contextual relevance, length of sentences, and amount of information represented by the frequency of adjectives. The sentence generator also has an additional phrasing skill. Three controlled lab experiments were conducted, where subjects were requested to read generated sentences and watch videos filmed about conversations between the robot and a person. The results implied that mechanisms effectively promote users' enjoyment and interests.

Original languageEnglish
Article number6924789
Pages (from-to)313-326
Number of pages14
JournalIEEE/ACM Transactions on Speech and Language Processing
Volume23
Issue number2
DOIs
Publication statusPublished - 2015 Feb 1

Fingerprint

conversation
sentences
Robots
Experiments
robot
robots
clocks
Person
generators
video
Robot
Generator
human being
experiment
Experiment
Skills

Keywords

  • Conversational robots
  • natural sentence generation
  • opinion generation
  • question answering

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Media Technology
  • Acoustics and Ultrasonics
  • Instrumentation
  • Linguistics and Language
  • Speech and Hearing

Cite this

Automatic expressive opinion sentence generation for enjoyable conversational systems. / Matsuyama, Yoichi; Saito, Akihiro; Fujie, Shinya; Kobayashi, Tetsunori.

In: IEEE/ACM Transactions on Speech and Language Processing, Vol. 23, No. 2, 6924789, 01.02.2015, p. 313-326.

Research output: Contribution to journalArticle

@article{140292c6484048e8b010f23bab84295b,
title = "Automatic expressive opinion sentence generation for enjoyable conversational systems",
abstract = "In terms of functional conversations, Grice's Maxim of Quantity suggests that responses should contain no more information than was explicitly asked for. However, in our daily conversations, more informative response skills are usually employed in order to hold enjoyable conversations with interlocutors. These responses are usually produced as forms of one's additional opinions, which usually contain their original viewpoints as well as novel means of expression, rather than simple and common responses characteristic of the general public. In this paper, we propose automatic expressive opinion sentence generation mechanisms for enjoyable conversational systems. The generated opinions are extracted from a large number of reviews on the web, and ranked in terms of contextual relevance, length of sentences, and amount of information represented by the frequency of adjectives. The sentence generator also has an additional phrasing skill. Three controlled lab experiments were conducted, where subjects were requested to read generated sentences and watch videos filmed about conversations between the robot and a person. The results implied that mechanisms effectively promote users' enjoyment and interests.",
keywords = "Conversational robots, natural sentence generation, opinion generation, question answering",
author = "Yoichi Matsuyama and Akihiro Saito and Shinya Fujie and Tetsunori Kobayashi",
year = "2015",
month = "2",
day = "1",
doi = "10.1109/TASLP.2014.2363589",
language = "English",
volume = "23",
pages = "313--326",
journal = "IEEE/ACM Transactions on Speech and Language Processing",
issn = "2329-9290",
publisher = "IEEE Advancing Technology for Humanity",
number = "2",

}

TY - JOUR

T1 - Automatic expressive opinion sentence generation for enjoyable conversational systems

AU - Matsuyama, Yoichi

AU - Saito, Akihiro

AU - Fujie, Shinya

AU - Kobayashi, Tetsunori

PY - 2015/2/1

Y1 - 2015/2/1

N2 - In terms of functional conversations, Grice's Maxim of Quantity suggests that responses should contain no more information than was explicitly asked for. However, in our daily conversations, more informative response skills are usually employed in order to hold enjoyable conversations with interlocutors. These responses are usually produced as forms of one's additional opinions, which usually contain their original viewpoints as well as novel means of expression, rather than simple and common responses characteristic of the general public. In this paper, we propose automatic expressive opinion sentence generation mechanisms for enjoyable conversational systems. The generated opinions are extracted from a large number of reviews on the web, and ranked in terms of contextual relevance, length of sentences, and amount of information represented by the frequency of adjectives. The sentence generator also has an additional phrasing skill. Three controlled lab experiments were conducted, where subjects were requested to read generated sentences and watch videos filmed about conversations between the robot and a person. The results implied that mechanisms effectively promote users' enjoyment and interests.

AB - In terms of functional conversations, Grice's Maxim of Quantity suggests that responses should contain no more information than was explicitly asked for. However, in our daily conversations, more informative response skills are usually employed in order to hold enjoyable conversations with interlocutors. These responses are usually produced as forms of one's additional opinions, which usually contain their original viewpoints as well as novel means of expression, rather than simple and common responses characteristic of the general public. In this paper, we propose automatic expressive opinion sentence generation mechanisms for enjoyable conversational systems. The generated opinions are extracted from a large number of reviews on the web, and ranked in terms of contextual relevance, length of sentences, and amount of information represented by the frequency of adjectives. The sentence generator also has an additional phrasing skill. Three controlled lab experiments were conducted, where subjects were requested to read generated sentences and watch videos filmed about conversations between the robot and a person. The results implied that mechanisms effectively promote users' enjoyment and interests.

KW - Conversational robots

KW - natural sentence generation

KW - opinion generation

KW - question answering

UR - http://www.scopus.com/inward/record.url?scp=84921668186&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84921668186&partnerID=8YFLogxK

U2 - 10.1109/TASLP.2014.2363589

DO - 10.1109/TASLP.2014.2363589

M3 - Article

AN - SCOPUS:84921668186

VL - 23

SP - 313

EP - 326

JO - IEEE/ACM Transactions on Speech and Language Processing

JF - IEEE/ACM Transactions on Speech and Language Processing

SN - 2329-9290

IS - 2

M1 - 6924789

ER -