Encoding emotional information for sequence-to-sequence response generation

Yin Hei Chan; Andrew Kwok Fai Lui

doi:10.1109/ICAIBD.2018.8396177

Encoding emotional information for sequence-to-sequence response generation

Yin Hei Chan, Andrew Kwok Fai Lui

Hong Kong Metropolitan University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Citations (Scopus)

Abstract

This paper introduces an alternative approach on embedding emotional information at the encoder stage of a sequence-to-sequence based emotional response generation. It explores different positioning and styles of the embedding, which represent associations of emotion with specific words or the whole sentence. The experiment was set up with standard dataset as well as dataset annotated with emotional classifiers. Preliminary results showed that this new approach should better represent sentence level emotional and work well with standard Recurrent Neural network (RNN) with Long Short Term Memory (LSTM) architecture.

Original language	English
Title of host publication	2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018
Pages	113-116
Number of pages	4
ISBN (Electronic)	9781538669877
DOIs	https://doi.org/10.1109/ICAIBD.2018.8396177
Publication status	Published - 25 Jun 2018
Event	2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018 - Chengdu, China Duration: 26 May 2018 → 28 May 2018

Publication series

Name	2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018

Conference

Conference	2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018
Country/Territory	China
City	Chengdu
Period	26/05/18 → 28/05/18

Keywords

Chatbots
Conversational agents
Emotional response
Encoder-decoder framework
Long short term memory
Recurrent neural network

Access to Document

10.1109/ICAIBD.2018.8396177

Cite this

@inproceedings{f612d624043e4fee8368a38f5e422118,

title = "Encoding emotional information for sequence-to-sequence response generation",

abstract = "This paper introduces an alternative approach on embedding emotional information at the encoder stage of a sequence-to-sequence based emotional response generation. It explores different positioning and styles of the embedding, which represent associations of emotion with specific words or the whole sentence. The experiment was set up with standard dataset as well as dataset annotated with emotional classifiers. Preliminary results showed that this new approach should better represent sentence level emotional and work well with standard Recurrent Neural network (RNN) with Long Short Term Memory (LSTM) architecture.",

keywords = "Chatbots, Conversational agents, Emotional response, Encoder-decoder framework, Long short term memory, Recurrent neural network",

author = "Chan, {Yin Hei} and Lui, {Andrew Kwok Fai}",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018 ; Conference date: 26-05-2018 Through 28-05-2018",

year = "2018",

month = jun,

day = "25",

doi = "10.1109/ICAIBD.2018.8396177",

language = "English",

series = "2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018",

pages = "113--116",

booktitle = "2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018",

}

Chan, YH & Lui, AKF 2018, Encoding emotional information for sequence-to-sequence response generation. in 2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018. 2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018, pp. 113-116, 2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018, Chengdu, China, 26/05/18. https://doi.org/10.1109/ICAIBD.2018.8396177

Encoding emotional information for sequence-to-sequence response generation. / Chan, Yin Hei; Lui, Andrew Kwok Fai.
2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018. 2018. p. 113-116 (2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Encoding emotional information for sequence-to-sequence response generation

AU - Chan, Yin Hei

AU - Lui, Andrew Kwok Fai

PY - 2018/6/25

Y1 - 2018/6/25

N2 - This paper introduces an alternative approach on embedding emotional information at the encoder stage of a sequence-to-sequence based emotional response generation. It explores different positioning and styles of the embedding, which represent associations of emotion with specific words or the whole sentence. The experiment was set up with standard dataset as well as dataset annotated with emotional classifiers. Preliminary results showed that this new approach should better represent sentence level emotional and work well with standard Recurrent Neural network (RNN) with Long Short Term Memory (LSTM) architecture.

AB - This paper introduces an alternative approach on embedding emotional information at the encoder stage of a sequence-to-sequence based emotional response generation. It explores different positioning and styles of the embedding, which represent associations of emotion with specific words or the whole sentence. The experiment was set up with standard dataset as well as dataset annotated with emotional classifiers. Preliminary results showed that this new approach should better represent sentence level emotional and work well with standard Recurrent Neural network (RNN) with Long Short Term Memory (LSTM) architecture.

KW - Chatbots

KW - Conversational agents

KW - Emotional response

KW - Encoder-decoder framework

KW - Long short term memory

KW - Recurrent neural network

UR - http://www.scopus.com/inward/record.url?scp=85050215111&partnerID=8YFLogxK

U2 - 10.1109/ICAIBD.2018.8396177

DO - 10.1109/ICAIBD.2018.8396177

M3 - Conference contribution

AN - SCOPUS:85050215111

T3 - 2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018

SP - 113

EP - 116

BT - 2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018

T2 - 2018 International Conference on Artificial Intelligence and Big Data, ICAIBD 2018

Y2 - 26 May 2018 through 28 May 2018

ER -

Encoding emotional information for sequence-to-sequence response generation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this