Word Embeddings and Their Use In Sentence Classification Tasks
This work demonstrates the effectiveness of pre-trained word embeddings for sentence classification, which is useful for NLP practitioners.
The paper explores word embeddings and their applications, then implements a convolutional neural network using pre-trained word vectors for sentence classification tasks, achieving state-of-the-art or comparable results.
This paper have two parts. In the first part we discuss word embeddings. We discuss the need for them, some of the methods to create them, and some of their interesting properties. We also compare them to image embeddings and see how word embedding and image embedding can be combined to perform different tasks. In the second part we implement a convolutional neural network trained on top of pre-trained word vectors. The network is used for several sentence-level classification tasks, and achieves state-of-art (or comparable) results, demonstrating the great power of pre-trainted word embeddings over random ones.