Abstract
A hashtag is a type of metadata tag used on social networks, such as Twitter and other microblogging services. Hashtags indicate the core idea of a microblog post and can help people to search for specific themes or content. However, not everyone tags their posts themselves. Therefore, the task of hashtag recommendation has received significant attention in recent years. To solve the task, a key problem is how to effectively represent the text of a microblog post in a way that its representation can be utilized for hashtag recommendation. We study two major kinds of text representation methods for hashtag recommendation, including shallow textual features and deep textual features learned by deep neural models. Most existing work tries to use deep neural networks to learn microblog post representation based on the semantic combination of words. In this paper, we propose to adopt Tree-LSTM to improve the representation by combining the syntactic structure and the semantic information of words. We conduct extensive experiments on two real world datasets. The experimental results show that deep neural models generally perform better than traditional methods. Specially, Tree-LSTM achieves significantly better results on hashtag recommendation than standard LSTM, with a 30% increase in F1-score, which indicates that it is promising to utilize syntactic structure in the task of hashtag recommendation.
Highlights
Social networking services (SNSs), such as Twitter, Facebook and Google+, have attracted millions of users to publish and share the most up-to-date information, emergent social events, and personal opinions [1]
We propose and compare various approaches to represent the text for hashtag recommendation, and extensive experiments demonstrate the powerful predictive ability of our deep neural network models on two real-world datasets
We study two major kinds of text representation methods for hashtag recommendation, including shallow textual features and deep textual features learned by deep neural models
Summary
Social networking services (SNSs), such as Twitter, Facebook and Google+, have attracted millions of users to publish and share the most up-to-date information, emergent social events, and personal opinions [1]. The massive amounts of information generated every day makes it difficult to discover the hidden information in that data To organize this information more accurately and effectively, many microblogging services allow users to create and use hashtags by placing the pound sign #, usually in front of a word or unspaced phrase in a post. We perform an experimental study of the short text representation methods in hashtag recommendation task. Memory network (LSTM) [12] in text classification, we propose using Tree-LSTM [13] to introduce syntactic structure while learning the representation of microblog post. We propose and compare various approaches to represent the text for hashtag recommendation, and extensive experiments demonstrate the powerful predictive ability of our deep neural network models on two real-world datasets
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have