Large-Scale Categorization of Japanese Product Titles Using Neural Attention Models

Abstract

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.

E-Commerce Product Categorization via Machine Translation

“…Each MT system in Section 4.2 tokenizes a product title into individual words and then outputs a root-to-leaf path one node category at a time, similar to how a translated sentence is generated one word at a time. The distribution of products across categories in both datasets is skewed toward the most popular categories as is usually the case in e-commerce domains [He and McAuley 2016;Xia et al 2017]. Figures 5 and 6 show the number of products in each category at the top level of the taxonomy tree (each vertical bar reflects the number of products in that category).…”

“…Kozareva [2015] uses a variety of features (e.g., n-grams, latent Dirichlet allocation topics [Blei et al 2003], and word2vec embeddings [Mikolov et al 2013]) in a multi-class algorithm. Both Ha et al [2016] and Xia et al [2017] use deep learning to learn a compact vector representation of the attributes of a product (e.g., product title, merchant ID, and product image), and use the representation to classify the product. They differ in terms of the kinds of deep learning model used.…”

Developing Data-Conscious Deep Learning Models for Product Classification

“…Xia, Y. et al [4] classified product categories using product title data in Japanese, which is an agglutinative language. An attention conventional neural network (ACNN) was proposed using a conventional CNN model and gradient boosted tree (GBT) [18].…”

“…This approach sets the weights of some particular layers appropriately and applies it in typical models such as a convolutional neural network (CNN) [1,2]. The second approach is to develop a technically new model [3,4].…”

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Papers

“…Deep learning-based methods have been widely used for the IC task. This includes the use of deep neural network models for item categorization in a hierarchical classifier structure which showed improved performance over conventional machine learning models (Cevahir and Murakami, 2016), as well as the use of an attention mechanism to identify words that are semantically highly correlated with the predicted categories and therefore can provide improved feature representations for a higher classification performance (Xia et al, 2017).…”