Article Contents
Article Contents

# Zinc ore supplier evaluation and recommendation method based on nonlinear adaptive online transfer learning

• * Corresponding author: Yonggang Li

The first author is supported by NNSFC grant 61973321

• Purchasing decisions determine the purchasing cost, which is the largest section of the production cost of zinc smelting enterprise(ZSE). An excellent supplier recommendation is significant for ZSE to reduce the cost. However, during the supplier recommendation process, the nonlinear demand feature of purchasing department varies with the production environment, and there are wrong samples that can affect the supplier recommendation effect. To handle these problems, the recommendation strategy based on a multiple-layer perceptron adaptive online transfer learning algorithm(AOTLMLP) are proposed. In this method, the original prediction function is modified based on MLP nonlinear projective function and adaptive loss function, which enables the AOTLMLP algorithm to tackle the nonlinear classification problems and efficiently follow the demand change of purchasing department, thereby improving the result of the recommendation. The performance of the AOTLMO algorithm is evaluated through a common dataset and a purchasing dataset from a zinc smelter that generated by a supplier evaluation model. It can be assumed that AOTLMLP can ignore the influence of wrong samples and provide an effective recommendation confronting the characteristic of zinc ore purchasing.

Mathematics Subject Classification: Primary: 68W27; Secondary: 90B06.

 Citation:

• Figure 1.  The framework of evaluation criterias for zinc ore suppliers

Figure 2.  The requirement change problem in supplier recommendation

Figure 3.  The structure of nonlinear prediction function based on MLP

Figure 4.  The update steps of algorithm

Figure 5.  Cumulative trainning error rate for different $\beta$, considering IMAGE dataset

Figure 6.  Cumulative trainning error rate for different $\beta$, considering Purchasing dataset

Figure 7.  Cumulative trainning error rate for different layer parameter, considering IMAGE dataset

Figure 8.  Cumulative trainning error rate for different layer parameter, considering Purchasing dataset

Figure 9.  Cumulative trainning error rate for different $\eta$, considering training purchasing dataset

Figure 10.  Cumulative trainning error rate for different $\eta$, considering testing purchasing dataset

Figure 11.  Recommendation accuracy for different $\eta$, for purchasing demand change

Figure 12.  Cumulative trainning error rate for different $\varphi$, considering training purchasing dataset

Figure 13.  Cumulative trainning error rate for different $\varphi$, considering testing purchasing dataset

Figure 14.  Recommendation accuracy for different $\varphi$, for purchasing demand change

Figure 15.  The convergence behaviors for different learning strategy, considering purchasing dataset

Figure 16.  The performance of four algorithms, considering purchasing demand change

Table 1.  Symbol reference table

 Symbol Paraphrase ${{\bf{x}}_t}$ Supplier feature vector ${y_t}$ Recommendation outcome ${\bf{v}}$ Linear previous demand feature vector ${{\bf{w}}_t}$ Linear present demand feature vector(time-varying) ${{\bf{v}}_\phi }$ Nonlinear previous demand feature matrix ${{\bf{w}}_{\phi t}}$ Nonlinear present demand feature matrix(time-varying) ${{\bf{z}}_t}$ The hidden layer node vector(time-varying) ${{\bf{z}}_{(j)t}}$ The jth hidden layer node vector(time-varying) ${{\bf{r}}_t}$ The weight vector for ReLU units(time-varying) ${{\bf{r}}_{(j)t}}$ The jth layer weight vector for ReLU units(time-varying) $\beta$ The restriction parameter $\varphi$ The preference parameter $\eta$ The transfer speed rate
•  [1] G. Akman, Evaluating suppliers to include green supplier development programs via fuzzy c-means and vikor methods, Computers & Industrial Engineering, 86 (2015), 69-82.  doi: 10.1016/j.cie.2014.10.013. [2] M. Balabanović and Y. Shoham, Fab: Content-based, collaborative recommendation, Communications of the ACM, 40 (1997), 66-72.  doi: 10.1145/245108.245124. [3] Z. Chen, Y. Jiang and Y. Zhao, A collaborative filtering recommendation algorithm based on user interest change and trust evaluation, International Journal of Digital Content Technology and its Applications, 4 (2010), 106-113. [4] J. Cheng, Y. Liu, H. Zhang, X. Wu and F. Chen, A new recommendation algorithm based on user's dynamic information in complex social network, Mathematical Problems in Engineering, 2015 (2015), Article ID 281629. doi: 10.1155/2015/281629. [5] G. Ditzler, M. Roveri, C. Alippi and R. Polikar, Learning in nonstationary environments: A survey, IEEE Computational Intelligence Magazine, 10 (2015), 12-25.  doi: 10.1109/MCI.2015.2471196. [6] M. M. Gaber, A. Zaslavsky and S. Krishnaswamy, Mining data streams: A review, ACM Sigmod Record, 34 (2005), 18-26.  doi: 10.1145/1083784.1083789. [7] J. Gama, I. Žliobaitė, A. Bifet, M. Pechenizkiy and A. Bouchachia, A survey on concept drift adaptation, ACM Computing Surveys (CSUR), 46 (2014), 1-37.  doi: 10.1145/2523813. [8] I. Gasmi, H. Seridi-Bouchelaghem, L. Hocine and B. Abdelkarim, Collaborative filtering recommendation based on dynamic changes of user interest, Intelligent Decision Technologies, 9 (2015), 271-281.  doi: 10.3233/IDT-140221. [9] T. Grubinger, G. C. Chasparis and T. Natschläger, Generalized online transfer learning for climate control in residential buildings, Energy and Buildings, 139 (2017), 63-71.  doi: 10.1016/j.enbuild.2016.12.074. [10] S. C. Hoi, D. Sahoo, J. Lu and P. Zhao, Online learning: A comprehensive survey, arXiv preprint, arXiv: 1802.02871. [11] B.-J. Hou, L. Zhang and Z.-H. Zhou, Prediction with unpredictable feature evolution, IEEE Transactions on Neural Networks and Learning Systems. [12] J. Jorge and R. Paredes, Passive-aggressive online learning with nonlinear embeddings, Pattern Recognition, 79 (2018), 162-171.  doi: 10.1016/j.patcog.2018.01.019. [13] Z. Kang, B. Yang, Z. Li and P. Wang, Otlamc: An online transfer learning algorithm for multi-class classification, Knowledge-Based Systems, 176 (2019), 133-146.  doi: 10.1016/j.knosys.2019.03.024. [14] J.-J. Kuo and Y.-J. Zhang, A library recommender system using interest change over time and matrix clustering, in International Conference on Asian Digital Libraries, Springer, 2012,259–268. doi: 10.1007/978-3-642-34752-8_32. [15] K. Lang, Newsweeder: Learning to filter netnews, in Machine Learning Proceedings 1995, 1995, 331–339. doi: 10.1016/B978-1-55860-377-6.50048-7. [16] J. Li, W. Qiu and W. Li, An improved k-means algorithm for supplier evaluation and recommendation of purchase and supply platform, in Journal of Physics: Conference Series, vol. 1650 (2020), 032165. doi: 10.1088/1742-6596/1650/3/032165. [17] Y.-H. Lin and L. Chang, An online transfer learning framework for time-varying distribution data prediction., [18] S. Luthra, K. Govindan, D. Kannan, S. K. Mangla and C. P. Garg, An integrated framework for sustainable supplier selection and evaluation in supply chains, Journal of Cleaner Production, 140 (2017), 1686-1698.  doi: 10.1016/j.jclepro.2016.09.078. [19] C. Lv, Y. Lu, X. Yan, W. Lu and H. Tan, Supplier recommendation based on knowledge graph embedding, in 2020 Management Science Informatization and Economic Innovation Development Conference (MSIEID), IEEE, 2020,514–518. doi: 10.1109/MSIEID52046.2020.00105. [20] A. Niyazov, E. Mikhailova and O. Egorova, Content-based music recommendation system, in 2021 29th Conference of Open Innovations Association (FRUCT), IEEE, 2021,274–279. doi: 10.23919/FRUCT52173.2021.9435533. [21] R. W. Saaty, The analytic hierarchy process-what it is and how it is used, Mathematical Modelling, 9 (1987), 161-176.  doi: 10.1016/0270-0255(87)90473-8. [22] U. Thakker, R. Patel and M. Shah, A comprehensive analysis on movie recommendation system employing collaborative filtering, Multimedia Tools and Applications, 1–26. [23] Q. Wu, X. Zhou, Y. Yan, H. Wu and H. Min, Online transfer learning by leveraging multiple source domains, Knowledge and Information Systems, 52 (2017), 687-707.  doi: 10.1007/s10115-016-1021-1. [24] P. Zhao, S. C. H. Hoi, J. Wang and B. Li, Online transfer learning, Artificial Intelligence, 216 (2014), 76-102.  doi: 10.1016/j.artint.2014.06.003. [25] I. Žliobaitė, Learning under concept drift: An overview, arXiv preprint, arXiv: 1010.4784.

Figures(16)

Tables(1)