Research on chat robot based on Seq2seq model

Jianjun Huang

doi:10.1117/12.2682987

16 June 2023 Research on chat robot based on Seq2seq model

Jianjun Huang

Proceedings Volume 12703, Sixth International Conference on Intelligent Computing, Communication, and Devices (ICCD 2023); 127032D (2023) https://doi.org/10.1117/12.2682987
Event: Sixth International Conference on Intelligent Computing, Communication, and Devices (ICCD 2023), 2023, Hong Kong, China

Abstract

More and more scholars put forward to apply deep learning to chat robots and realize the dialogue processing of chat robots through semantic analysis and emotional dialogue characteristics of deep learning. In actual chat, all kinds of sentences will exist, but there are some uncertainties in the logic of sentences and the part of speech of words, and some questions have different meanings of the same word. In order to carry on the dialogue, the robot must have strong semantic disambiguation ability and remember the answered questions. In this article, an intelligent chat robot semantic matching algorithm based on Seq2seq is proposed, and under certain attribute matching rules, the semantic similarity of a concept at the next level pointed by the attribute is found, and the similarity of each concept is calculated, so that the similarity of all concepts at all levels is weighted and integrated to obtain the semantic similarity between ontologies. The experimental results show that the dialogue generation mechanism designed this time has higher recognition rate than the traditional mechanism, and can correctly identify the dialogue content to make accurate answers, which provides technical support for the development of intelligent chat robots.

Citation Download Citation

Jianjun Huang "Research on chat robot based on Seq2seq model", Proc. SPIE 12703, Sixth International Conference on Intelligent Computing, Communication, and Devices (ICCD 2023), 127032D (16 June 2023); https://doi.org/10.1117/12.2682987

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available