Paper
16 June 2023 Research on semantic matching algorithm of BERT intelligent question answering system
Ming Gao, Tianyao Ji, Luliang Zhang, Wuchen Tang
Author Affiliations +
Proceedings Volume 12703, Sixth International Conference on Intelligent Computing, Communication, and Devices (ICCD 2023); 127032A (2023) https://doi.org/10.1117/12.2683003
Event: Sixth International Conference on Intelligent Computing, Communication, and Devices (ICCD 2023), 2023, Hong Kong, China
Abstract
With the rapid growth and widespread popularity of the Internet, people's access to information is diversified and convenient. The existing text representation methods have some problems, such as insufficient semantic information extraction, high dimension of representation model, and high complexity of model construction. In this paper, a semantic matching algorithm of intelligent question answering system based on BERT is proposed, and the semantic similarity of a concept at the next level pointed by the attribute is found under certain attribute matching rules. Finally, the concept similarity method is recursively called to calculate the similarity of each concept, so that the similarity of all concepts at all levels is weighted and integrated to obtain the semantic similarity between ontologies. The simulation results show that after the system is deployed, the ideal effect can be obtained by comparing the accuracy and response time of text selection. This shows that the improved method proposed in this paper can effectively improve the performance of BERT model, and then effectively reduce the model parameters and accelerate the pre-training speed under the same performance.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ming Gao, Tianyao Ji, Luliang Zhang, and Wuchen Tang "Research on semantic matching algorithm of BERT intelligent question answering system", Proc. SPIE 12703, Sixth International Conference on Intelligent Computing, Communication, and Devices (ICCD 2023), 127032A (16 June 2023); https://doi.org/10.1117/12.2683003
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Semantics

Intelligence systems

Data modeling

Data conversion

Detection and tracking algorithms

Genetic algorithms

Performance modeling

Back to Top