Collaborative Preference Learning: A Case Study
Okan Tunalı, Ahmet Tuğrul Bayrak · 2020 4th International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT) · 2020
In the scope of marketing operations, the transformation of transactional data into customer-item preference indicators is a critical process. As in the case of fast food sector, one needs to estimate preferences or ratings based on that sales data. The ratings can then be used to estimate customer similarities and utilised in recommendation systems. In this study, we elaborately explain the process of sales data transformation into preferences and propose a trend concerning collaborative filtering method. In the experiments, we highlight the domain specific data processing requirements and demonstrate that our method makes significanly better predictions.