 |
    |
Lagrange Multiplier Method for Multi-campaign Assignment Problem
Yong-Hyuk Kim and Byung-Ro Moon
School of Computer Science & Engineering, Seoul National University, Shillim-dong, Kwanak-gu, Seoul, 151-742 Koreayhdfly@soar.snu.ac.kr
moon@soar.snu.ac.kr
Abstract. It is crucial to maximize marketing efficiency and customer satisfaction in personalized marketing. In this paper, we raise the multiple recommendation problem which occurs when performing several personalized campaigns simultaneously. We formulate the multi-campaign assignment problem to solve this issue and propose methods for solving the problem. The most notable element is the Lagrange multiplier method. Roughly speaking, Lagrange multiplier reduces problem complexity with a minor impact on optimality. However, it is not easy to find Lagrange multipliers in exact accord with problem constraints. We use a genetic algorithm for finding optimal Lagrange multipliers. Through the experiments, we verify the effectiveness of the problem formulation and our genetic approach.
LNCS 3103, p. 1065 ff.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2004