Abstract
Nonnegative tensor factorization (NTF) plays a crucial role in extracting latent factors and predicting future sales from purchase data consisting of user and item attributes. However, the increase in these attributes leads to tensor data becoming sparse, causing a reduction in decomposition accuracy. For example, when there are numerous combinations of unavailable item genres and prices, the purchase history data becomes sparse and follows a distribution where all its elements are zero. To address this issue, we propose a novel NTF method assuming zero-inflated Poisson (ZIP) distribution based on Expectation-Maximization (EM) algorithm. This enables us to effectively handle sparsity in high-dimensional multiway data and identify combinations of user and item attributes that are potentially not likely to be purchased. We verified the effectiveness of the proposed approach through numerical experiments using real-world e-commerce data. The results showed our proposed ZIP model outperforms existing methods in both in-sample and out-of-sample experiments. Moreover, the proposed method qualitatively demonstrated the effectiveness of handling sparsity.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Lee, D., Seung, H.S.: Algorithms for non-negative matrix factorization. In: Advances in Neural Information Processing Systems, vol. 13. MIT Press (2000)
Rafailidis, D., Daras, P.: The TFC model: tensor factorization and tag clustering for item recommendation in social tagging systems. IEEE Trans. Syst., Man Cybern.: Syst. 43(3), 673–688 (2012)
Taneja, A., Arora, A.: Cross domain recommendation using multidimensional tensor factorization. Expert Syst. Appl. 92, 304–316 (2018)
Yin, H., et al.: SPTF: a scalable probabilistic tensor factorization model for semantic-aware behavior prediction. In: 2017 IEEE International Conference on Data Mining (ICDM), pp. 585–594. IEEE (2017)
Chou, S.-Y., Jang, J.-S.R., Yang, Y.-H.: Fast tensor factorization for large-scale context-aware recommendation from implicit feedback. IEEE Trans. Big Data 6(1), 201–208 (2018)
Narita, A., Hayashi, K., Tomioka, R., Kashima, H.: Tensor factorization using auxiliary information. Data Min. Knowl. Disc. 25, 298–324 (2012)
Zhou, P., Lu, C., Lin, Z., Zhang, C.: Tensor factorization for low-rank tensor completion. IEEE Trans. Image Process. 27(3), 1152–1163 (2017)
Lambert, D.: Zero-inflated Poisson regression, with an application to defects in manufacturing. Technometrics 34(1), 1–14 (1992)
Coxe, S., West, S.G., Aiken, L.S.: The analysis of count data: a gentle introduction to Poisson regression and its alternatives. J. Pers. Assess. 91(2), 121–136 (2009)
Hiroyasu, A., Hiroshi, Y.: A non-negative matrix factorization model based on the zero-inflated Tweedie distribution. Comput. Statistics 32(2), 475–499 (2017)
Cichocki, A., Rafal, Z., Amari, S.: Nonnegative matrix and tensor factorization [lecture notes]. IEEE Signal Process. Mag. 25(1), 142–145 (2008)
Mørup, M.: Applications of tensor (multiway array) factorizations and decompositions in data mining. Wiley Interdisc. Rev.: Data Min. Knowl. Discovery 1(1), 24–40 (2011)
Chen, Y., He, W., Yokoya, N., Huang, T.-Z.: Hyperspectral image restoration using weighted group sparsity-regularized low-rank tensor decomposition. IEEE Trans. Cybern. 50(8), 3556–3570 (2020)
Li, S., Dian, R., Fang, L., Bioucas-Dias, J.M.: Fusing hyperspectral and multispectral images via coupled sparse tensor factorization. IEEE Trans. Image Process. 27(8), 4118–4130 (2018)
Mitsufuji, Y., Roebel, A.: Sound source separation based on non-negative tensor factorization incorporating spatial cue as prior knowledge. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 71–75 (2013)
Gligorijević, V., Panagakis, Y., Zafeiriou, S.: Non-negative matrix factorizations for multiplex network analysis. IEEE Trans. Pattern Anal. Mach. Intell. 41(4), 928–940 (2018)
Jain, P., Oh, S.: Provable tensor factorization with missing data. In: Advances in Neural Information Processing Systems, vol. 27. Curran Associates, Inc., (2014)
Lopez, O., Lehoucq, R., Dunlavy, D.: Zero-truncated Poisson tensor decomposition for sparse count data. Sandia National Lab.(SNL-NM), Albuquerque, NM (United States), Technical Report (2022)
Yu, H.F., Rao, N., Dhillon, I.S.: Temporal regularized matrix factorization for high-dimensional time series prediction. In: Advances in Neural Information Processing Systems, vol. 29. Curran Associates, Inc., (2016)
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. Roy. Stat. Soc.: Ser. B (Methodological) 39(10), 1–22 (1977)
Kossaifi, J., Panagakis, Y., Anandkumar, A., Pantic, M.: TensorLy: tensor learning in python. J. Mach. Learn. Res. 20, 1–6 (2016)
Daly, F., Gaunt, R.E.: The Conway-Maxwell-Poisson distribution: distributional theory and approximation. ALEA 13, 635–658 (2016)
Gladys, F.L., Barriga, D.C.: The zero-inflated Conway-Maxwell-Poisson distribution: Bayesian inference, regression modeling and influence diagnostic. Stat. Methodol. 21, 23–34 (2014)
Shmueli, G., Minka, T.P., Kadane, J.B., Borle, S., Boatwright, P.: A useful distribution for fitting discrete data: revival of the Conway-Maxwell-Poisson distribution. J. Roy. Stat. Soc.: Ser. C: Appl. Stat. 54(1), 127–142 (2005)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendices
A Computational Time
We compare the seconds per step of the above experiments to evaluate computational time. Figure 5 (a) shows the average computational time of 10 trial runs for the four factorization models. The initial values are uniformly sampled across all models for each trial. Due to the complexity of parameters and update algorithms, our proposed method resulted in the slowest computational time.
Additionally, we compared the computational time of implementations using our JAX + JIT compiler method with Tensorly in Fig. 5 (b). Both of these methods assume a normal distribution on the tensor and use the same algorithm. Our proposed algorithm significantly mitigates the increase in computational time with the increase in rank compared to the alternative.
B ZICMP Distribution Model
We demonstrate the Zero-Inflated Conway-Maxwell-Poisson (ZICMP) distribution tensor factorization model, which is a modified version of the Conway-Maxwell-Poisson (CMP) distribution in place of the Poisson distribution [22,23,24]. This enables to represent over and under dispersion which was not able by the ZIP distribution. When random variable y follows this distribution, the probability mass function is defined as follows:
p is the mixture ratio and \(\lambda \) is the expected value of the ZICMP distribution. \(\nu \) is a nonnegative real number and controls the variance. For \(\nu = 1\), its distribution aligns with the Poisson distribution.
In contrast to the ZIP model, here we update \(\boldsymbol{A}^{(1)},\dots , \boldsymbol{A}^{(n)}\) and \(\nu \) using Newton’s method at each step of factor matrices updating phase in Algorithm 1. A magnitude of Newton step can be derived by taking the first and second derivatives of the log-likelihood function with respect to \(a^{(k)}_{i_kr}\) and \(\nu \).
We present the results of the numerical experiments for tensor factorization assuming the ZICMP distribution in Table 4. Our algorithm encountered overflow with large tensor sizes or element values. To address this, we reduced the tensor size to a 5th-mode tensor by eliminating the region mode.
As a result, RMSPE outperformed the other algorithms, but not in other evaluation metrics. While it can be assumed that the ZICMP distribution improved the expressiveness of the distribution, the fitting to sparsity did not work.
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Mizutani, K. et al. (2024). Zero-Inflated Poisson Tensor Factorization for Sparse Purchase Data in E-Commerce Markets. In: Sheu, SH. (eds) Industrial Engineering and Applications – Europe. ICIEA-EU 2024. Lecture Notes in Business Information Processing, vol 507. Springer, Cham. https://doi.org/10.1007/978-3-031-58113-7_14
Download citation
DOI: https://doi.org/10.1007/978-3-031-58113-7_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-58112-0
Online ISBN: 978-3-031-58113-7
eBook Packages: Computer ScienceComputer Science (R0)