Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System

Jordan J. Bird, Anikó Ekárt, Diego R. Faria

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper we propose an approach to a chatbot software that is able to learn from interaction via text messaging between human-bot and bot-bot. The bot listens to a user and decides whether or not it knows how to reply to the message accurately based on current knowledge, otherwise it will set about to learn a meaningful response to the message through pattern matching based on its previous experience. Similar methods are used to detect offensive messages, and are proved to be effective at overcoming the issues that other chatbots have experienced in the open domain. A philosophy of giving preference to too much censorship rather than too little is employed given the failure of Microsoft Tay. In this work, a layered approach is devised to conduct each process, and leave the architecture open to improvement with more advanced methods in the future. Preliminary results show an improvement over time in which the bot learns more responses. A novel approach of message simplification is added to the bot’s architecture, the results suggest that the algorithm has a substantial improvement on the bot’s conversational performance at a factor of three.
Original languageEnglish
Title of host publicationUK Workshop on Computational Intelligence
PublisherSpringer
Chapter15
Pages179-190
Volume840
ISBN (Electronic)978-3-319-97982-3
ISBN (Print)978-3-319-97981-6
DOIs
Publication statusE-pub ahead of print - 11 Aug 2018
EventUKCI'18: 18th Annual UK Workshop on Computational Intelligence - Nottingham, United Kingdom
Duration: 5 Sep 20187 Sep 2018

Publication series

NameAdvances in Computational Intelligence Systems
Volume840
ISSN (Print)2194-5357
ISSN (Electronic)2194-5365

Conference

ConferenceUKCI'18: 18th Annual UK Workshop on Computational Intelligence
CountryUnited Kingdom
CityNottingham
Period5/09/187/09/18

Fingerprint

Text messaging
Pattern matching

Cite this

Bird, J. J., Ekárt, A., & Faria, D. R. (2018). Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System. In UK Workshop on Computational Intelligence (Vol. 840, pp. 179-190). (Advances in Computational Intelligence Systems; Vol. 840). Springer. https://doi.org/10.1007/978-3-319-97982-3_15
Bird, Jordan J. ; Ekárt, Anikó ; Faria, Diego R. / Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System. UK Workshop on Computational Intelligence. Vol. 840 Springer, 2018. pp. 179-190 (Advances in Computational Intelligence Systems).
@inproceedings{6b0e5b8f4e3a4c4f9423b509e218346e,
title = "Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System",
abstract = "In this paper we propose an approach to a chatbot software that is able to learn from interaction via text messaging between human-bot and bot-bot. The bot listens to a user and decides whether or not it knows how to reply to the message accurately based on current knowledge, otherwise it will set about to learn a meaningful response to the message through pattern matching based on its previous experience. Similar methods are used to detect offensive messages, and are proved to be effective at overcoming the issues that other chatbots have experienced in the open domain. A philosophy of giving preference to too much censorship rather than too little is employed given the failure of Microsoft Tay. In this work, a layered approach is devised to conduct each process, and leave the architecture open to improvement with more advanced methods in the future. Preliminary results show an improvement over time in which the bot learns more responses. A novel approach of message simplification is added to the bot’s architecture, the results suggest that the algorithm has a substantial improvement on the bot’s conversational performance at a factor of three.",
author = "Bird, {Jordan J.} and Anik{\'o} Ek{\'a}rt and Faria, {Diego R.}",
year = "2018",
month = "8",
day = "11",
doi = "10.1007/978-3-319-97982-3_15",
language = "English",
isbn = "978-3-319-97981-6",
volume = "840",
series = "Advances in Computational Intelligence Systems",
publisher = "Springer",
pages = "179--190",
booktitle = "UK Workshop on Computational Intelligence",
address = "Germany",

}

Bird, JJ, Ekárt, A & Faria, DR 2018, Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System. in UK Workshop on Computational Intelligence. vol. 840, Advances in Computational Intelligence Systems, vol. 840, Springer, pp. 179-190, UKCI'18: 18th Annual UK Workshop on Computational Intelligence, Nottingham, United Kingdom, 5/09/18. https://doi.org/10.1007/978-3-319-97982-3_15

Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System. / Bird, Jordan J.; Ekárt, Anikó; Faria, Diego R.

UK Workshop on Computational Intelligence. Vol. 840 Springer, 2018. p. 179-190 (Advances in Computational Intelligence Systems; Vol. 840).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System

AU - Bird, Jordan J.

AU - Ekárt, Anikó

AU - Faria, Diego R.

PY - 2018/8/11

Y1 - 2018/8/11

N2 - In this paper we propose an approach to a chatbot software that is able to learn from interaction via text messaging between human-bot and bot-bot. The bot listens to a user and decides whether or not it knows how to reply to the message accurately based on current knowledge, otherwise it will set about to learn a meaningful response to the message through pattern matching based on its previous experience. Similar methods are used to detect offensive messages, and are proved to be effective at overcoming the issues that other chatbots have experienced in the open domain. A philosophy of giving preference to too much censorship rather than too little is employed given the failure of Microsoft Tay. In this work, a layered approach is devised to conduct each process, and leave the architecture open to improvement with more advanced methods in the future. Preliminary results show an improvement over time in which the bot learns more responses. A novel approach of message simplification is added to the bot’s architecture, the results suggest that the algorithm has a substantial improvement on the bot’s conversational performance at a factor of three.

AB - In this paper we propose an approach to a chatbot software that is able to learn from interaction via text messaging between human-bot and bot-bot. The bot listens to a user and decides whether or not it knows how to reply to the message accurately based on current knowledge, otherwise it will set about to learn a meaningful response to the message through pattern matching based on its previous experience. Similar methods are used to detect offensive messages, and are proved to be effective at overcoming the issues that other chatbots have experienced in the open domain. A philosophy of giving preference to too much censorship rather than too little is employed given the failure of Microsoft Tay. In this work, a layered approach is devised to conduct each process, and leave the architecture open to improvement with more advanced methods in the future. Preliminary results show an improvement over time in which the bot learns more responses. A novel approach of message simplification is added to the bot’s architecture, the results suggest that the algorithm has a substantial improvement on the bot’s conversational performance at a factor of three.

UR - http://link.springer.com/10.1007/978-3-319-97982-3_15

U2 - 10.1007/978-3-319-97982-3_15

DO - 10.1007/978-3-319-97982-3_15

M3 - Conference contribution

SN - 978-3-319-97981-6

VL - 840

T3 - Advances in Computational Intelligence Systems

SP - 179

EP - 190

BT - UK Workshop on Computational Intelligence

PB - Springer

ER -

Bird JJ, Ekárt A, Faria DR. Learning from Interaction: An Intelligent Networked-Based Human-Bot and Bot-Bot Chatbot System. In UK Workshop on Computational Intelligence. Vol. 840. Springer. 2018. p. 179-190. (Advances in Computational Intelligence Systems). https://doi.org/10.1007/978-3-319-97982-3_15