Assessing new correlation-based collaborative filtering approaches for binary market basket data

Binary market basket data are common in marketing settings. Pearson correlation-based approaches have applied to this kind of data in the past. This research assessed the principles of this approach research identifies some related problems. By resolving the problems, I develop new Pearson correlation-based approaches that use separated terms and separated terms with proportions. The experimental results show that the approaches perform better than the existing ones for market basket data in terms of top-N accuracy.