We propose a variant of Convolutional Neural Network (CNN) models, the Attention CNN (ACNN); for large-scale categorization of millions of Japanese items into thirty-five product categories. Compared to a state-of-the-art Gradient Boosted Tree (GBT) classifier, the proposed model reduces training time from three weeks to three days while maintaining more than 96% accuracy. Additionally, our proposed model characterizes products by imputing attentive focus on word tokens in a language agnostic way. The attention words have been observed to be semantically highly correlated with the predicted categories and give us a choice of automatic feature extraction for downstream processing.

No abstract

The Transformer has proven to be a powerful feature extraction method and has gained widespread adoption in natural language processing (NLP). In this paper we propose a multimodal item categorization (MIC) system solely based on the Transformer for both text and image processing. On a multimodal product data set collected from a Japanese ecommerce giant, we tested a new image classification model based on the Transformer and investigated different ways of fusing bi-modal information. Our experimental results on real industry data showed that the Transformerbased image classifier has performance on par with ResNet-based classifiers and is four times faster to train. Furthermore, a cross-modal attention layer was found to be critical for the MIC system to achieve performance gains over text-only and image-only models.

In this paper, we propose a task for quality evaluation of disputing argument. In order to understand the disputation behavior, we propose three sub-tasks, detecting disagreement hierarchy, refutation method and argumentation strategy respectively. We first manually labeled a real dataset collected from an online debating forum. The dataset includes 45 disputing argument pairs. The annotation scheme is developed by three NLP researchers via annotating all the argument pairs in the dataset. Two under-graduate students are then trained to annotate the same dataset. We report annotation results from both groups. Then, another larger dataset was annotated and we show analysis of the correlation between disputing quality and different disputation behaviors.

No abstract