I am new to Neo4j and have been trying out with the PageRank algorithm for a side project of mine which involves recommending products. I understood that PageRank gives a static ranking of the nodes(product) and I wish to make it "dynamic" by factoring in the frequency of visit to each node(product) over the past week/month.
I have tried to factor in the frequency of visit by updating the weights of the relationships pointing to the node(product) with the count of visits to the node over the past week and then running the PageRank algorithm to update each node's weight. Is this the correct way to do it? And if so, what are the benefits of using PageRank over simply just recommending the products based on the frequency of visit over the past week?