Traditional Chinese Medicine Text Similarity Calculation Model Based on the Bidirectional Temporal Siamese Network

<table class="algorithm-group"><tr><td><table class="algorithm" id="alg1"><tr><td>(1)</td><td>Input: dataset D</td></tr><tr><td>(2)</td><td>Process:</td></tr><tr><td>(4)</td><td>The dataset D is divided into training set and test set according to 7 : 3</td></tr><tr><td>(5)</td><td>Import the pretrained language model ERNIE to obtain the dynamic character vector. The similarity calculation model is obtained through the training set.</td></tr><tr><td>(6)</td><td>Forward-LEFT = LSTM-LEFT (vector); forward-RIGHT = LSTM-RIGHT (vector);</td></tr><tr><td>(7)</td><td>Backward-LEFT = LSTM-LEFT (vector); backward-RIGHT = LSTM-RIGHT (vector)</td></tr><tr><td>(8)</td><td>H-LEFT = connect (Forward-LEFT, Backward-LEFT); H-RIGHT = connect (Forward-RIGHT, Backward-RIGHT);</td></tr><tr><td>(9)</td><td>Siamese (H-LEFT, H-RIGHT)</td></tr><tr><td>(10)</td><td>Return similarity calculation model</td></tr><tr><td>(11)</td><td>Use the test set to get various evaluations</td></tr><tr><td>(12)</td><td>End</td></tr></table></td></tr></table>

<div> Bidirectional temporal Siamese network (BTSN).</div>

Evidence-Based Complementary and Alternative Medicine

alg1

Algorithm 1

Algorithm 1: Traditional Chinese Medicine Text Similarity Calculation Model Based on the Bidirectional Temporal Siamese Network 