Multichannel Variable-Size Convolution for Sentence Classification

TL;DR

This paper introduces MVCNN, a CNN architecture that combines diverse pretrained embeddings and variable-size filters to improve sentence classification, achieving state-of-the-art results across multiple datasets.

Contribution

The paper presents a novel CNN architecture that integrates multiple pretrained embeddings and multigranular phrase features for enhanced sentence classification.

Findings

MVCNN achieves state-of-the-art performance on four benchmark tasks.

Pretraining MVCNN is essential for optimal performance.

Combining diverse embeddings and variable-size filters improves classification accuracy.

Abstract

We propose MVCNN, a convolution neural network (CNN) architecture for sentence classification. It (i) combines diverse versions of pretrained word embeddings and (ii) extracts features of multigranular phrases with variable-size convolution filters. We also show that pretraining MVCNN is critical for good performance. MVCNN achieves state-of-the-art performance on four tasks: on small-scale binary, small-scale multi-class and largescale Twitter sentiment prediction and on subjectivity classification.