Abstract
Markup languages such as RDFa and Microdata have been widely used by e-shops to embed structured product data, as evidence has shown that they improve click-through rates for e-shops and potentially increases their sales. While e-shops often embed certain categorisation information in their product data in order to improve their products’ visibility to product search and aggregator services, such site-specific product category labels are highly inconsistent and unusable across websites. This work studies the task of automatically classifying products into a universal categorisation taxonomy, using their markup data published on the Web. Using three new neural network models adapted based on previous work, we analyse the effect of different kinds of product markup data on this task, and show that: (1) despite the highly heterogeneous nature of the site-specific categories, they can be used as very effective features - even only by themselves - for the classification task; and (2) our best performing model can significantly improve state of the art on this task by up to 9.6% points in macro-average F1.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
- 2.
- 3.
- 4.
This separator is chosen as it is the most commonly used in the dataset (described in Sect. 4).
- 5.
- 6.
- 7.
sg: http://schema.org/.
- 8.
The literature has mostly used name + other features, which we also do in this work. Also, as we shall show in the results, among n, d, and c alone, d generally performs the worst. So we do not report results based on description + other features.
- 9.
- 10.
This is an assumption based on our observation, as we have been unable to confirm this with the authors despite our efforts. However, we assume this is the truth as our macro-average results are the closest, while our micro-and weighted macro-average results are significantly higher (in the range between 70 and 90).
- 11.
References
Akritidis, L., Fevgas, A., Bozanis, P.: Effective unsupervised matching of product titles with k-combinations and permutations. In: IEEE 30th International Conference on Tools with Artificial Intelligence, pp. 213–220 (2018)
Altınel, B., CanGaniz, M.: Semantic text classification: a survey of past and recent advances. Inf. Process. Manag. 54(6), 1129–1153 (2018)
Cevahir, A., Murakami, K.: Large-scale multi-class and hierarchical product categorization for an e-commerce giant. In: Proceedings of the 26th International Conference on Computational Linguistics (COLING): Technical Papers, pp. 525–535. The COLING 2016 Organizing Committee (2016)
Chavaltada, C., Pasupa, K., Hardoon, D.R.: A comparative study of machine learning techniques for automatic product categorisation. In: Cong, F., Leung, A., Wei, Q. (eds.) ISNN 2017. LNCS, vol. 10261, pp. 10–17. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-59072-1_2
Ding, Y., et al.: Automated classification of product data in e-commerce. In: Proceedings of the Business Information Systems Conference (2002)
Gupta, V., Karnick, H., Bansal, A., Jhala, P.: Product classification in e-commerce using distributional semantics. In: Proceedings of COLING 2016: Technical Papers, pp. 536–546. The COLING 2016 Organizing Committee (2016)
Ha, J.W., Pyo, H., Kim, J.: Large-scale item categorization in e-commerce using multiple recurrent neural networks. In: Proceedings of the International Conference on Knowledge Discovery and Data Mining (KDD), pp. 107–115. ACM (2016)
Kim, Y.: Convolutional neural networks for sentence classification. In: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1746–1751. Association for Computational Linguistics (ACL) (2014)
Kim, Y., Lee, T., Chun, J., Lee, S.: Modified Naïve Bayes classifier for e-catalog classification. In: Lee, J., Shim, J., Lee, S., Bussler, C., Shim, S. (eds.) DEECS 2006. LNCS, vol. 4055, pp. 246–257. Springer, Heidelberg (2006). https://doi.org/10.1007/11780397_20
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Proceedings of the 3rd International Conference for Learning Representations (2015)
Kozareva, Z.: Everyone likes shopping! multi-class product categorization for e-commerce. In: Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), pp. 1329–1333. ACL (2015)
Lee, H., Yoon, Y.: Engineering doc2vec for automatic classification of product descriptions on O2O applications. Electron. Commer. Res. 18(3), 433–456 (2018)
Li, M.Y., Kok, S., Tan, L.: Don’t classify, translate: multi-level e-commerce product categorization via machine translation. CoRR. http://arxiv.org/abs/1812.05774
Lin, Y.C., Das, P., Datta, A.: Overview of the SIGIR 2018 eCom Rakuten data challenge. In: eCom Data Challenge at SIGIR 2018. ACM (2018)
McCaffrey, J.D.: Why you should use cross-entropy error instead of classification error or mean squared error for neural network classifier training. https://jamesmccaffrey.wordpress.com. Accessed Jan 2018
Meusel, R., Primpeli, A., Meilicke, C., Paulheim, H., Bizer, C.: Exploiting microdata annotations to consistently categorize product offers at web scale. In: Stuckenschmidt, H., Jannach, D. (eds.) EC-Web 2015. LNBIP, vol. 239, pp. 83–99. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-27729-5_7
Ristoski, P., Petrovski, P., Mika, P., Paulheim, H.: A machine learning approach for product matching and categorization. Seman. Web 9(5), 707–728 (2018)
Sebastiani, F.: Machine learning in automated text categorization. ACM Comput. Surv. 34(1), 1–47 (2002)
Xia, Y., Levine, A., Das, P., Di Fabbrizio, G., Shinzato, K., Datta, A.: Large-scale categorization of Japanese product titles using neural attention models. In: Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics (EACL), vol. 2, Short Papers, pp. 663–668. Association for Computational Linguistics (2017)
Yang, Z., Yang, D., Dyer, C., He, X., Smola, A., Hovy, E.: Hierarchical attention networks for document classification. In: Proceedings of NAACL 2016, pp. 1480–1489. Association for Computational Linguistics (2016)
Zhou, P., Qi, Z., Zheng, S., Xu, J., Bao, H., Xu, B.: Text classification improved by integrating bidirectional LSTM with two-dimensional max pooling. In: Proceedings of the 26th International Conference on Computational Linguistics: Technical Papers, pp. 3485–3495. The COLING 2016 Organizing Committee (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, Z., Paramita, M. (2019). Product Classification Using Microdata Annotations. In: Ghidini, C., et al. The Semantic Web – ISWC 2019. ISWC 2019. Lecture Notes in Computer Science(), vol 11778. Springer, Cham. https://doi.org/10.1007/978-3-030-30793-6_41
Download citation
DOI: https://doi.org/10.1007/978-3-030-30793-6_41
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-30792-9
Online ISBN: 978-3-030-30793-6
eBook Packages: Computer ScienceComputer Science (R0)