Market Analysis Using a Combination of Bayesian Networks and Description Logics
Phillip M. Yelland · 1999
The work described in this paper was inspired by a problem increasingly vexatious to many businesses confronting the ever-diminishing life cycles of modern products---viz., that of predicting characteristics (such as overall demand, segmentation, etc.) of markets facing new product introductions. A framework is proposed that allows the market parameters of new products to be derived by analogy with those of old ones. To do so, the framework combines the capabilities of Bayesian networks [14] and description logics [16]. The paper commences with an exposition of the problem that motivates the work. There follow brief descriptions of Bayesian networks and description logics in their unalloyed state, and a discussion of issues surrounding their combination. The combined system is presented, along with an account of its formal details and an inference procedure for entailment. A sample application of the framework is given. The paper concludes by comparing the proposed framework with related existing systems and by suggesting possible courses for future development.