Multi-Objective Group Discovery on the Social Web

摘要: We are interested in discovering user groups from collaborative rating datasets of the form $$\langle i, u, s\rangle $$, where $$i \in \mathcal{I}$$, $$u \mathcal{U}$$, and s is integer that u has assigned to item i. Each a set attributes help find labeled such as young computer scientists France American female designers. formalize problem finding whose quality optimized multiple dimensions show it NP-Complete. develop $$\alpha $$-MOMRI, an $$-approximation algorithm, h-MOMRI, heuristic-based for multi-objective optimization high groups. Our extensive experiments on real social Web examine performance our algorithms report cases $$-MOMRI h-MOMRI useful.