An End-to-End Rumor Detection Model Based on Feature Aggregation

Algorithm 1

Temporal text feature construction.

	Input:  =  temporal text data related to event
	Output: structured sequence of temporal text related to event
	batched_post  0;
	sort_by_time ();
	for in do
	batch_size //Calculate the batch size.
	start  batched_post;
	doc ;
	batch_  tf_idf (doc, IDF, K)//Calculate the top-k words in TF-IDF value.
	embedded_batch   word_embedding (batch ); //Obtaining word vector.
	batched_post  batched_post + batch_size;
	return embedded_batch