Open problems in CF Scalability problem Cr is computationally expensive. Fastest published algorithms(nearest-neighbor) are n Any indexing method for speeding up? Has received relatively little attention a References in cf http://www.cs.sfu.ca/cc/470/gyAng/lectures/cf rethtm
16 Open Problems in CF ◼ “Scalability Problem” ◼ CF is computationally expensive. Fastest published algorithms (nearest-neighbor) are n 2 . ◼ Any indexing method for speeding up? ◼ Has received relatively little attention. ◼ References in CF: ◼ http://www.cs.sfu.ca/CC/470/qyang/lectures/cf ref.htm
References P Domingos and M. richardson, Mining the Network value of customers, Proceedings of the Seventh International conference on Knowledge Discovery and data Mining(pp 57-66), 2001 San francisco, ca: acm press 17
17 References ◼ P. Domingos and M. Richardson, Mining the Network Value of Customers, Proceedings of the Seventh International Conference on Knowledge Discovery and Data Mining (pp. 57-66), 2001. San Francisco, CA: ACM Press
Motivation Network value is ignored(direct marketing) Examples Market to Affected (under the network effect) expecte profit sected ofit arketed High expected profit 18
18 Motivation ◼ Network value is ignored (Direct marketing). ◼ Examples: Market to Affected (under the network effect) Low expected profit High expected profit High expected profit Marketed
Some successful Case a Hotmail Grew from o to 12 million users in 18 months Each email include a promotional url of it ICQ Expand quickly First appear, user addicted to it Depend it to contact with friend 19
19 Some Successful Case ◼ Hotmail ◼ Grew from 0 to 12 million users in 18 months ◼ Each email include a promotional URL of it. ◼ ICQ ◼ Expand quickly ◼ First appear, user addicted to it ◼ Depend it to contact with friend