Skip to main content
Have a personal or library account? Click to login
Research on Improved Dual Channel Medical Short Text Intention Recognition Algorithm Cover

Research on Improved Dual Channel Medical Short Text Intention Recognition Algorithm

By: ,   and    
Open Access
|Mar 2024

Figures & Tables

Figure 1.

Architecture diagram of AB-CNN-BGRU-att model

Figure 2.

ALBERT model structure

Figure 3.

GRU network structure

Figure 4.

BiGRU network structure

Figure 5.

Improved TextCNN model structure

TABLE I.

experimental dataset

NameTraining SetTest SetValidation SetCategoryTotal
KUAKE-QIC6931199419551110880
THUCNews_Title180000100001000010200000
Figure 6.

Model validation results

Figure 7.

Comparison of Network Time

TABLE II.

comparison between bigru-att and bilstm-att

Network LayerAverage DurationTotal DurationAcc%F1%
BiGRU-att2286.9s45738s90.8390.64
BiLSTM-att2422.55s48451s90.4590.41
TABLE III.

comparison of experimental results

ModelAcc%Pre%Recall%F1%
SAttBiGRU96.1696.2096.1696.17
Self-Attention-CNN94.8594.8994.8594.85
BiGRU-MCNN95.4395.4595.4395.43
MC-AttCNN-AttBiGRU95.9395.9895.9395.93
TABLE IV.

results of ablation experiment

ModelAcc%Pre%Recall%F1%
TextCNN89.9689.9089.9689.90
Improved TextCNN94.8594.8994.8594.85
BiGRU-att94.0094.1794.0094.90
AB-CNN-BGRU-att96.6896.6896.6796.67
Language: English
Page range: 82 - 89
Published on: Mar 16, 2024
Published by: Xi’an Technological University
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2024 Chao Wang, Yongyong Sun, Fei Xu, published by Xi’an Technological University
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.