Estimating frequency of change
ACM Transactions on Internet Technology (TOIT)
Estimation of internet file-access/modification rates from indirect data
ACM Transactions on Modeling and Computer Simulation (TOMACS)
Effective change detection using sampling
VLDB '02 Proceedings of the 28th international conference on Very Large Data Bases
Web science: an interdisciplinary approach to understanding the web
Communications of the ACM - Web science
Hi-index | 0.00 |
In the Internet-based economy, the (relative) transparency of e-markets and increasing online market dynamics call for more responsive and encompassing approaches towards the monitoring of markets and competitors. Accordingly, this paper proposes to observe continuously a preselected set of e-commerce Web channels, or online portals, to gather a comprehensive as possible picture of market dynamics. In so doing, a historical market data repository is accumulated based on an adaptive scheme of harvesting Web data online in order to provide dynamic information about both market structure and prices. A description of the proposed estimator for online data sampling based on observed (price) change frequencies is given. Numerical simulations highlight the virtues of the proposed adaptive estimator compared to established Web page change frequency estimators, even more so in case of considering constraints on (observation) resources. As an example, the methodology is applied to the online hotel room booking market.