This is the authors' abstract. We don't add key points for this paper.
The successfulness reached by numerous neural network models for encoding word embedding has conducted approaches for encoding vector representation for sentences, paragraphs or even micro-blogs.. Meanwhile, the hashtag is a keyword that denotes the topic of a tweet. Hashtags supply significant information for several text mining tasks, for instance sentiment classification, news analysis, etc. Hence, an appropriate hashtag recommender system is needed to assist users in choosing relevant hashtags for their tweets. Therefore, we present a hashtags recommender method to encode the tweet vector-based representation by using a long short-term memory recurrent neural network. Specifically, we first learn words vector-based representation by training the Skip gram model to generate embeddings of tweets. Then we apply the density-based spatial clustering to gather similar tweets into homogeneous clusters. Subsequently, we recommend the k top highest results of tweets to the user, via calculating their co-occurrence with the nearest clusters centers to a given tweet. Our experiments on a real data set show appropriate results.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Ben-Lhachemi et al. (2019) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: