Tag Recommendation by Word-Level Tag Sequence Modeling
This addresses tag recommendation for users on platforms like Zhihu, but it is incremental as it adapts existing sequence-to-sequence techniques to this task.
The paper tackled tag recommendation by framing it as a word-based text generation problem using a sequence-to-sequence model, and the result showed that the proposed model outperformed state-of-the-art text classification methods on Zhihu datasets.
In this paper, we transform tag recommendation into a word-based text generation problem and introduce a sequence-to-sequence model. The model inherits the advantages of LSTM-based encoder for sequential modeling and attention-based decoder with local positional encodings for learning relations globally. Experimental results on Zhihu datasets illustrate the proposed model outperforms other state-of-the-art text classification based methods.