Recommendation of Items with Short Life Span
October 20, 2016
11:10 am - 11:30 am
GRB 330 A
Track:
Artificial Intelligence
Type: Presentation
Level:
Advanced
We propose a recommendation system designed for items with short life span. The difficulties are frequent cold start of new items and sparsity problem. Our approach is to combine four techniques, namely, user based collaborative filtering, profile matching on keywords, purchase pattern analysis of categories and sales volume. Test results show that our system out-performs widely used algorithms.
Speaker(s)
, SAP |