From: Cross-modality representation learning from transformer for hashtag prediction
Refs. | Year | Name | Features | Techniques | Recommendation |
---|---|---|---|---|---|
[9] | 2016 | TAB_LSTM | Text | Topical Attention-Based LSTM model that merge local hidden representations with global topic vectors. | General |
[11] | 2018 | HRMF | Text | Hashtag recommendation method based on multi-features(hashtag, user, text) of microblogs. To find similar users they proposed a new topic model User-Hashtag Topic Model Based on Short Text Expansion (UHTME). | Personalized |
[12] | 2018 | Hashtag2Vec | Text | Explored the multiple relations of hashtag-tweet, tweet-word, word-word, and hashtag-hashtag relationships based on the hierarchical heterogeneous network. | General |
[13] | 2019 | DeepTagRec | Text | A neural network model for tag recommendation on Stack Overflow, Quora, etc that leverages both the textual content (title and body) of the questions and the user-tag network. | Personalized |
[14] | 2019 | TCAN | Text | Topical Co-Attention Network which learns the content representation with BiLSTM and constructs the topical word matrix and combines them with the co-attention mechanism. | General |
[15] | 2019 | PLSTM | Text | Parallel Long Short-term Memory based on current post contents and the post history representation. | Personalized |
[16] | 2018 | STR | Text | Semantically Enhanced Tag Recommendation approach that recommends the tags through semantics learning of both tags and questions on Stack Overflow. | General |
[22] | 2017 | CNN-PerMLP | Image | For personalized hashtag recommendation consider the user’s preferences as well as visual information for image tags recommendation. | Personalized |
[23] | 2018 | A-NIH | Image | Attention-based Neural Image Hashtag network for sequence relationships between images and hashtags process with Inception V3 and GRU | General |
[25] | 2020 | VDNN-ARM | Image | A voting deep neural network with associative rules mining approach. | General |
[26] | 2020 | – | Image | A user conditional joint embedding model, it first extracted the visual representation for each image and then computed the vectorial representation of each image-hashtags into pairs. | Personalized |
[27] | 2017 | CoA | Image, text | A deep learning framework that incorporates textual(LSTM) and visual(CNN) features of multimodal tweets with co-attention model. | General |
[29] | 2019 | MACON | Image, text | Memory augmented co-attention network which learns the image and text fractures with user tagging habits. | Personalized |
[30] | 2020 | AMNN | Image, text | A sequence generation Attention-based Multimodal Neural Network that extracts the features of images and texts and incorporates them into the sequence-to-sequence GRU model for hashtag recommendation. | General |
[33] | 2021 | CACNet | Image, text | VGG and weighted average Word2Vec based Cross-Active Connection model for Image-Text Feature Fusion. | General |
[34] | 2022 | TweetEmbd_Net | Image, text | Explore the problem of jointly modeling tweet components (image, text, hashtags, user, and, location) in a common embedding space. | General |