A Wine Consumption Prediction Model Based on L-DAGLSSVM
With the increasing demand of wine consumption, the marketing of wine consumption is expanding. In this paper, we do a research about the decision behavior of Chinese wine consumers in order to grasp the consumption demand of wine at different prices better. We acquire 774 questionnaires finally, and the 528 of which are valid. According to the consumption prices, we divide wine consumers into three types. Then we propose a multi-class classification method named L-DAGLSSVM for constructing prediction model of consumption types, which is based on LDA and the directed acyclic graph least squares support vector machine (DAGLSSVM). The numerical experiments demonstrate that our algorithm gains better performance compared with other algorithms. And the prediction model plays an important role in commercial fields that it can provide an effective reference for the wine production, purchase and marketing strategies etc.
KeywordsLSSVM The decision directed acyclic graph (DDAG) LDA Prediction model of consumption types
- 1.Naiyang, D., & Yingjie, T. (2009). Support vector machine: theory, algorithm and development. Beijing: Science Press (in Chinese).Google Scholar
- 2.Cristianini, N., & Shawe-Taylor, J. (2004). An introduction to support vector machines and other kernel-based learning methods. Beijing: Publishing House of Electronics Industry.Google Scholar
- 4.Kressel, B. U. (1999). Pairwise classification and support vector machines, advances in kernel methods: Support vector learning. Cambridge: MIT Press.Google Scholar
- 6.Yang, X., Yu, Q., He, L., et al. (2013). The one-against-all partition based binary tree support vector machine algorithms for multi-class classification. Neural Computing, 113, 1–7.Google Scholar
- 7.Platt, J. C., Cristianini, N., & Shawe-Taylor, J. (2000). Large margin DAGs for multiclass classification. Advances in Neural Information Processing Systems, 12(3), 547–553.Google Scholar