Having clearly established the superior quality of item-based algorithms
over the user-based ones, we focus on the scalability challenges. As we
discussed earlier, item-based similarity is more static and allows us to
precompute the item neighborhood. This precomputation of the model has
certain performance benefits. To make the system even more scalable we looked
into the sensitivity of the model size and then looked into the impact of model
size on the response time and throughput.