Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
480423 | European Journal of Operational Research | 2016 | 11 Pages |
•Multitype multiway data are introduced for cross-selling recommendations.•A collaborative ensemble learning framework is developed.•The multi-kernel support tensor machine is proposed.•The method can select features from large sparse multitype multiway data.
Cross-selling is an integral component of customer relationship management. Using relevant information to improve customer response rate is a challenging task in cross-selling recommendations. Incorporating multitype multiway customer behavioral, including related product, similar customer and historical promotion, data into cross-selling models is helpful in improving the classification performance. Customer behavioral data can be represented by multiple high-order tensors. Most existing supervised tensor learning methods cannot directly deal with heterogeneous and sparse multiway data in cross-selling recommendations. In this study, a novel collaborative ensemble learning method, multi-kernel support tensor machine (MK-STM), is proposed for classification in cross-selling recommendations using multitype multiway customer behavioral data. The MK-STM can also perform feature selections from large sparse multitype multiway data. Computational experiments are conducted using two databases. The experimental results show that the MK-STM has better performance than existing ensemble learning, supervised tensor learning and other commonly used recommendation methods for cross-selling recommendations.