eGitty

Discover The Most Popular Algorithms

How to Incorporate External Knowledge in Recurrent Neural Networks (RNNs) for Text Classification

In this article, we will introduce three methods to incorporate external knowledge in recurrent neural networks (RNNs) for text classification.

Why use external knowlege in text classification?

External knowlege, such as sentiment lexicons, may provide some useful information for text classification. We can incorporate them into our model to imporve the performance.

How to incorporate?

We can find three methods in paper: Attention-based Conditioning Methods for External Knowledge Integration

They are:

Attentional Concatenation

Attentional Concatenation

Attentional Feature-based Gating

Attentional Feature-based Gating

Attentional Affine Transformation

Attentional Affine Transformation

The architecture of these three methods is:

How to Incorporate External Knowledge in Recurrent Neural Networks (RNNs) for Text Classification

Which method is best?

From this paper, we can find Attentional Feature-based Gating is best.

Attentional Feature-based Gating Result

How to implement these three methods?

We an find an implementation in here.

Leave a Reply

Your email address will not be published. Required fields are marked *