A Multigranularity Text Driven Named Entity Recognition CGAN Model for Traditional Chinese Medicine Literatures

Algorithm 1

The training process. Minibatch stochastic gradient descent training of MT-CGAN. The number of steps to apply to the discriminator, , is a hyperparameter. We used . is the fusion of text’s multigranularity features, seed of entity distribution features and attention data for different types of TCM texts.

	Input: The fusion of text’s multi-granularity features .
	Output: The Loss function value .
(1)	For number of training iterations do
(2)	for k steps do
(3)	Sample minibatch of samples from prior
(4)	Sample minibatch of example from data generating distribution
(5)	Update the discriminator by ascending its stochastic gradient:
(6)
(7)	end for
(8)	Sample minibatch of samples from prior
(9)	Update the discriminator by ascending its stochastic gradient:
(10)
(11)	end for
(12)	The gradient-based updates can use any standard gradient-based learning rule. We used Adam in our experiments.