|
Title:
|
A SYNTHETIC AND COMPUTATIONAL LANGUAGE MODEL FOR INTERACTIVE DIALOGUE SYSTEM |
|
Author(s):
|
Zhao Han, Fuji Ren, Duoqian Miao |
|
ISBN:
|
978-989-8533-39-5 |
|
Editors:
|
Ajith P. Abraham, Antonio Palma dos Reis and Jörg Roth |
|
Year:
|
2015 |
|
Edition:
|
Single |
|
Keywords:
|
Affective Computing, Natural Language Understanding, Dialogue System, Language Model, Perspective Coordinate |
|
Type:
|
Full Paper |
|
First Page:
|
73 |
|
Last Page:
|
80 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
To satisfy the requirements of our Human-Robot Interactive Dialogue System, we propose a novel language model called Synthetic and Computational Language Model (SCLM), which can synthetically represent the dialogue paragraph information at multiple aspects such as semantics and emotion, and can also adapt to the flexible addition of modification in spoken language dialogue. Using the proposed model a given paragraph from both the output of the dialogue robot and the input of the user can be represented like math expressions composed of an ordered list of vectors with matrices and word-operators. The vectors contain most of the semantic information of the paragraph and the matrices contain most of the affective information. The most outstanding advantage of the model is that all the transformation or modification can be represented by adding multiplying matrices flexibly, and can increase or decrease the computing fineness according to operating environment through many ways such as changing the dimension of the matrices. The experiment shows that the model can meet the system needs well. |
|
|
|
|
|
|
