Abstract
In microblogging services, authors can use hashtags to mark keywords or topics in microblogs. Many live social media applications (e.g., microblog retrieval, classification) can gain great benefit from these manually labeled tags. However, only a small partition of microblogs contains hashtags. Moreover, many microblog posts contain not only textual content but also images. These visual resources also provide valuable information that is not included in the textual content. To recommend hashtags for these multimodal microblogs, in this work, we propose a novel generative method incorporating textual and visual information to solve the task. Experimental results on the data collected from real world microblogging services demonstrate that the proposed method outperforms state-of-the-art methods using either textual or visual information. The relative improvement of the proposed method over the textual only method is more than 17.1% in F1-score.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call