Large-scale Multi-label Text Classification - Revisiting Neural Networks
This work addresses multi-label text classification for researchers and practitioners, offering an incremental improvement by applying existing deep learning techniques to a known bottleneck.
The authors tackled the problem of large-scale multi-label text classification by revisiting neural networks, showing that a simple neural network with advanced training techniques like ReLU, dropout, and AdaGrad performs as well as or outperforms state-of-the-art methods on six diverse datasets.
Neural networks have recently been proposed for multi-label classification because they are able to capture and model label dependencies in the output layer. In this work, we investigate limitations of BP-MLL, a neural network (NN) architecture that aims at minimizing pairwise ranking error. Instead, we propose to use a comparably simple NN approach with recently proposed learning techniques for large-scale multi-label text classification tasks. In particular, we show that BP-MLL's ranking loss minimization can be efficiently and effectively replaced with the commonly used cross entropy error function, and demonstrate that several advances in neural network training that have been developed in the realm of deep learning can be effectively employed in this setting. Our experimental results show that simple NN models equipped with advanced techniques such as rectified linear units, dropout, and AdaGrad perform as well as or even outperform state-of-the-art approaches on six large-scale textual datasets with diverse characteristics.