Abstract
We describe a software system, called just enough delivery (JED), for optimising single-copy newspaper sales, based on a combination of neural and Bayesian technology. The prediction model is a huge feedforward neural network, in which each output corresponds to the sales prediction for a single outlet. Input-to-hidden weights are shared between outlets. The hidden-to-output weights are specific to each outlet, but linked through the introduction of priors. All weights and hyperparameters can be inferred using (empirical) Bayesian inference. The system has been tested on data for several different newspapers and magazines. Consistent performance improvements of 1 to 3% more sales with the same total amount of deliveries have been obtained.
Similar content being viewed by others
References
Ragg T, Menzel W, Baum W and Wigbers M (2002) Bayesian learning for sales rate prediction for thousands of retailers. Neurocomputing 43:127–144
MacKay D (1995) Probable networks and plausible predictions—a review of practical Bayesian methods for supervised neural networks. Network 6:469–505
Heskes T (1998) Solving a huge number of similar tasks: a combination of multi-task learning and a hierarchical Bayesian approach. In: Proceedings of the International Conference on Machine Learning, Morgan Kaufmann, San Mateo, CA
Heskes T (2000) Empirical Bayes for learning to learn. In: Langley P (ed) Proceedings of the Seventeenth International Conference on Machine Learning, Morgan Kaufmann, San Francisco, CA
Bakker B, Heskes T (2003) Task clustering and gating for Bayesian multitask learning. J Mach Learn Res 4:83–99
Caruana R (1997) Multitask learning. Mach Learn 28:41–75
Baxter J (1997) A Bayesian/information theoretic model of learning to learn via multiple task sampling. Mach Learn 28:7–39
Cadez I, Ganey S and Smyth P (2000) A general probabilistic framework for clustering individuals. Technical report, University of California, Irvine, CA
Bryk A, Raudenbusch S (1992) Hierarchical linear models. Sage, Newbury Park, UK
Robert C (1994) The Bayesian choice: a decision-theoretic motivation. Springer, Berlin Heidelberg New York
Wolpert D (1993) On the use of evidence in neural networks. In: Hanson S, Cowan J and Giles L (eds) Advances in neural information processing systems 5, Morgan Kaufmann, San Mateo, CA
Dempster A, Laird N and Rubin D (1977) Maximum likelihood from incomplete data via the EM algorithm. J Roy Stat Soc B 39:1–38
West M, Harrison J (eds) (1977) Bayesian forecasting and dynamic models. Springer, Berlin Heidelberg New York
Gamerman D, Migon H (1993) Dynamic hierarchical models. J Roy Stat Soc B 55:629–642
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Heskes, T., Spanjers, JJ., Bakker, B. et al. Optimising newspaper sales using neural-Bayesian technology. Neural Comput & Applic 12, 212–219 (2003). https://doi.org/10.1007/s00521-003-0384-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-003-0384-x