Tayfun Keskin

PUBLICATIONS

Published

Tomak, K., and Keskin, T. "Exploring the trade-off between immediate gratification and delayed network externalities in the consumption of information goods," European Journal of Operational Research (187:3), Jun 2008, pp 887-902.
ABSTRACT: Our motivation for this paper originates from recent advances in consumption and payment technologies available for mobile commerce. Ubiquitous computing is enabling consumers worldwide to reach digital content and services whenever and wherever they request it. This capability further fuels “impatience” in consumption of such information goods. We model such consumer behavior using a hyperbolic discounting approach. A subset of these products, especially software, also inherit delayed network externalities as part of their consumption characteristics. This builds a tension between decision to consume now or to expedite consumption. We build a stylized model to assess the impact of immediate gratification on the profit maximizing behavior of a monopolist firm which produces an information good with network externalities. We find that serving “impatient” consumers is always profitable for a monopolist. For lower levels of network externalities, the monopolist can increase first period and decrease second period prices in equilibrium. As network externalities effect increases, prices converge to the traditional market (with exponential discounters) levels.

Accepted

David Pardoe, Peter Stone, Maytal Saar-Tsechansky, Tayfun Keskin and Kerem Tomak “Adaptive Auction Mechanism Design and the Incorporation of Prior Knowledge,” Informs Journal of Computing, (2009)
Online auctions are key selling mechanisms for many corporations and individuals alike. As the auction choices for the sellers increase, the need for an automated support system to sort the parameter space and find optimal auction settings also increase. We propose a machine learning approach to seller decision support in online B2C and sponsored search auctions. Our simulation results suggest that auctioneers can attain more revenues by using learning algorithms.

Work in Progress

Tayfun Keskin, Collaborative Information Technology Moderation in Dynamic Teamwork with Team Member Departure, Dissertation (est. Jan 2010)
Collaborative information technologies have become an indispensable part of teamwork especially in project and system development. They offered cost reduction, effective means of communication and reliable archival along with many abilities that previously considered impractical. There is considerable work done on the impact of IT tools but both academics and practitioners continue being skeptic for the benefits of collaborative information technologies.