多屬性效用模式與基因演算法在網際網路媒體行銷組合最佳化之決策輔助

The Optimization of Web Media Mix：The Hybrid Approach of MAUT and GA

丁雅芳(Ya Fang Ting)

指導教授 ： 邱昭彰

元智大學/管理學院/資訊研究所/碩士(1998年)

摘要

隨著網際網路﹝Internet﹞的風行，上網人口急速增加，網路使用族群的迅速成長，使得網路日漸成為一個提供廣告、行銷、利益分配以及資訊服務的管道，並被視為一種做商業經營工具的新媒體，由於各類網站數量急遽成長，傳統單一網站的廣告方式已漸不符合潮流與效益，正如同傳統媒體組合計畫的方式，網路廣告亦能組合方式進行組合規劃。因此如何在面對眾多的網站廣告代理商或者廣告主在分配網路廣告預算時，在可能的網站中做一最適化之資源分配應是一個重要的研究議題。
本研究利用多屬性效用理論與基因演算法之技術建立「網際網路媒體行銷組合最佳化之決策輔助」，藉由多屬性效用理論推導出廣告主主觀之偏好值，結合客觀的廣告效果衡量指標，利用基因演算法的演化特性，尋找出最大廣告效益的網路媒體組合計畫，提供廣告主網路媒體組合計畫之決策參考。

關鍵字

多屬性效用理論 ； 遺傳基因演算法 ； 網路廣告 ； 媒體組合

並列摘要

As internet applications popularity and web users increase drastically. Internet has gradually become an effective channel for advertisement, marketing, profit shares and information services as well as new media for engaging business. Because there are variety of web sites that provide different types of service, conventional styles of advertising contents in a single web-site have been no longer effective in budgetary and timing concerns. Just like the way of portfolio analysis in traditional media plan, web advertisement can also be done in a similar approach to appropriately allocate the budget in order to achieve optimum advertising effectiveness.
This research adopts MAU and GA techniques to construct a decision model to support web advertisement portfolio decision. With the aid of MAU, the subjective preference can be more systematically and consistently integrated with objective measurement data. Further, based on the derived integrated information strength of searching capability, the optimum advertisement plan can be effectively determined by exploiting the GAs.

並列關鍵字

Multi-attribute Utility Theory ； Genetic Algorithms ； Web Advertising ； Media Mix

參考文獻

Bayne, K.M. The Internet Marketing Plan, John Wiley & Sons, Canada,1997.

Caldwell, C. & Johnston, V.S., “Tracking a Criminal Suspect Through ‘Face-Space’ with a Genetic Algorithms.” In Proceedings of the Fourth International Conference on Genetic Algorithms, See Belew and Booker,1991.

Chan, C.L. & Chung,C.C., “A Method Combining MAU and Fuzzy Logic for Cooperative Decision Making”, Computer and Industrial Engineering,35(1-2),pp.291-294,1998.

Chen, C.,Vempati, V. Aljaber, N., “An Application of Genetic Algorithm for Flow Shop Problems.” European Journal of Operational Research, 80, pp. 389-396,1995.

Clemen, R.T., Making Hard Decisions. Duxbury Press, 1991.