id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-152740-ln2dbqn2	Najafi, Ali	ComStreamClust: A communicative text clustering approach to topic detection in streaming data	2020-10-11		.txt	text/plain	4504	258	64	In order to tackle the aforementioned topic detection problem, we propose a communicative text clustering approach for tweet clustering, which has been applied on the COVID-19 and FA CUP datasets, which is described with greater details in Section 3. The obtained results provide confirmatory evidence that the proposed approach is effective and superior to the existing ones in topic detection on Twitter data. al [13] propose a model based on the universal sentence encoder [14] and transformers [15] to detect the main topics of tweets regarding the COVID-19 pandemic. The problem tackled in this paper can be formally defined as follows: Each data-point is assumed as a quadruple (id, t, ts, s), such that id is a unique value as the identification number; t is the text with at most 280 characters; ts is the timestamp of the tweet including its arrival date and time; and s is the subject of the tweet which is not known in advance. This paper proposed a new topic detection approach using stream clustering on Twitter data.	./cache/cord-152740-ln2dbqn2.txt	./txt/cord-152740-ln2dbqn2.txt