KNN vs. DNN: auto chatbot

Qiushi Xu

doi:10.1117/12.2655749

7 October 2022 KNN vs. DNN: auto chatbot

Qiushi Xu

Proceedings Volume 12344, International Conference on Intelligent and Human-Computer Interaction Technology (IHCIT 2022); 123440Y (2022) https://doi.org/10.1117/12.2655749
Event: International Conference on Intelligent and Human-Computer Interaction Technology (IHCIT 2022), 2022, Zhuhai, China

Abstract

Machine learning has played a very important role these days. A very common area related to daily life is auto chatbots, such as Apple Siri, Google Home, Mi Xiaoai, and others. These technologies all use deep learning as the foundation to achieve the goal of communicating with humans. Engineers and computer scientists put a massive amount of effort into trying to improve the quality of chatbots by improving the models that drive this feature. From previous research, KNN (K-nearest neighbors) and DNN (Deep Neural Network) are two widely used models in the machine learning area. To find out which is more efficient to manage auto-chat while understanding deeper how chatbots actually manage to recognize human language and quickly come up with corresponding answers, the two learning models were applied to a self-made chatbot. By comparing the effectiveness of applying K-nearest neighbors and Deep Neural Network, the paper finds that KNN runs faster and takes less space since it is a comparatively simpler algorithm. In terms of correctness, in this experiment, both algorithms turned out to have a similar percentage of correctness. This might be caused by a comparatively small dataset.

Citation Download Citation

Qiushi Xu "KNN vs. DNN: auto chatbot", Proc. SPIE 12344, International Conference on Intelligent and Human-Computer Interaction Technology (IHCIT 2022), 123440Y (7 October 2022); https://doi.org/10.1117/12.2655749

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