Learning Sentence-to-Hashtags Semantic Mapping for Hashtag Recommendation on Microblogs

Abstract

The growing use of microblogging platforms is generating a huge amount of posts that need effective methods to be classified and searched. In Twitter and other social media platforms, hashtags are exploited by users to facilitate the search, categorization, and spread of posts. Choosing the appropriate hashtags for a post is not always easy for users, and therefore posts are often published without hashtags or with hashtags not well defined. To deal with this issue, we propose a new model, called HASHET ( HAshtag recommendation using Sentence-to-Hashtag Embedding Translation ), aimed at suggesting a relevant set of hashtags for a given post. HASHET is based on two independent latent spaces for embedding the text of a post and the hashtags it contains. A mapping process based on a multi-layer perceptron is then used for learning a translation from the semantic features of the text to the latent representation of its hashtags. We evaluated the effectiveness of two language representation models for sentence embedding and tested different search strategies for semantic expansion, finding out that the combined use of BERT ( Bidirectional Encoder Representation from Transformer ) and a global expansion strategy leads to the best recommendation results. HASHET has been evaluated on two real-world case studies related to the 2016 United States presidential election and COVID-19 pandemic. The results reveal the effectiveness of HASHET in predicting one or more correct hashtags, with an average F -score up to 0.82 and a recommendation hit-rate up to 0.92. Our approach has been compared to the most relevant techniques used in the literature ( generative models , unsupervised models, and attention-based supervised models ) by achieving up to 15% improvement in F -score for the hashtag recommendation task and 9% for the topic discovery task.

Hashtag recommendation for enhancing the popularity of social media posts

“…Cantini et al. ( 2021 ) used a BERT-based hashtag recommendation methodology that maps semantic features of social media posts in the latent representation of their hashtags. Li et al.…”

Analyzing voter behavior on social media during the 2020 US presidential election campaign

“… 2020 ), while the fourth follows the approach to topic detection in social data proposed in Cantini et al. ( 2021 ). IOM-NN is an opinion mining technique aimed at discovering the political polarization of social media users during election campaigns characterized by the competition of political factions.…”

“…This step is aimed at identifying the main politically-related discussion topics characterizing the 2020 US election campaign, by following the unsupervised approach used in Cantini et al. ( 2021 ). As a first step, a Word2Vec model is trained on the entire corpus of tweets, in order to get the latent representation of hashtags and words in a 150-dimensional vector space.…”

“…As an example, a hashtag recommendation model can be used for enriching the information content of the analyzed data, since keyword-based approaches like IOM-NN are strongly dependent on the availability of consistent hashtags in social media posts (Cantini et al. 2021 ).…”

“…Firstly, we extracted the main discussion topics characterizing the 2020 US election campaign by leveraging the unsupervised approach proposed in Cantini et al. ( 2021 ), which relies on the density-based clustering of the latent representation of trending hashtags. Afterward, in order to achieve a more accurate representation of social media conversation, we studied the weekly evolution of the detected topics, which is useful to understand how online discussion evolves over time.…”