![](https://cdn.statically.io/img/biblio.ugent.be/static/images/publication-blank.png)
Predicting website audience demographics for web advertising targeting using multi-website clickstream data
- Author
- Koen De Bock (UGent) and Dirk Van den Poel (UGent)
- Organization
- Abstract
- Several recent studies have explored the virtues of behavioral targeting and personalization for online advertising. In this paper, we add to this literature by proposing a cost-effective methodology for the prediction of demographic website visitor profiles that can be used for web advertising targeting purposes. The methodology involves the transformation of website visitors' clickstream patterns to a set of features and the training of Random Forest classifiers that generate predictions for gender, age, level of education and occupation category. These demographic predictions can support online advertisement targeting (i) as an additional input in personalized advertising or behavioral targeting, or (ii) as an input for aggregated demographic website visitor profiles that support marketing managers in selecting websites and achieving an optimal correspondence between target groups and website audience composition. The proposed methodology is validated using data from a Belgian web metrics company. The results reveal that Random Forests demonstrate superior classification performance over a set of benchmark algorithms. Further, the ability of the model set to generate representative demographic website audience profiles is assessed. The stability of the models over time is demonstrated using out-of-period data.
- Keywords
- out-of-period validation, Random Forests, ensemble classification, clickstream analysis, web advertising, web user profiling, demographic targeting, demographic prediction, AREA, SITES, EXPOSURE, BEHAVIOR, SELECTION, CLASSIFICATION, ONLINE, ROC CURVE, RANDOM FORESTS, BANNER ADVERTISEMENTS
Downloads
-
(...).pdf
- full text
- |
- UGent only
- |
- |
- 308.17 KB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-967442
- MLA
- De Bock, Koen, and Dirk Van den Poel. “Predicting Website Audience Demographics for Web Advertising Targeting Using Multi-Website Clickstream Data.” FUNDAMENTA INFORMATICAE, vol. 98, no. 1, 2010, pp. 49–70, doi:10.3233/FI-2010-216.
- APA
- De Bock, K., & Van den Poel, D. (2010). Predicting website audience demographics for web advertising targeting using multi-website clickstream data. FUNDAMENTA INFORMATICAE, 98(1), 49–70. https://doi.org/10.3233/FI-2010-216
- Chicago author-date
- De Bock, Koen, and Dirk Van den Poel. 2010. “Predicting Website Audience Demographics for Web Advertising Targeting Using Multi-Website Clickstream Data.” FUNDAMENTA INFORMATICAE 98 (1): 49–70. https://doi.org/10.3233/FI-2010-216.
- Chicago author-date (all authors)
- De Bock, Koen, and Dirk Van den Poel. 2010. “Predicting Website Audience Demographics for Web Advertising Targeting Using Multi-Website Clickstream Data.” FUNDAMENTA INFORMATICAE 98 (1): 49–70. doi:10.3233/FI-2010-216.
- Vancouver
- 1.De Bock K, Van den Poel D. Predicting website audience demographics for web advertising targeting using multi-website clickstream data. FUNDAMENTA INFORMATICAE. 2010;98(1):49–70.
- IEEE
- [1]K. De Bock and D. Van den Poel, “Predicting website audience demographics for web advertising targeting using multi-website clickstream data,” FUNDAMENTA INFORMATICAE, vol. 98, no. 1, pp. 49–70, 2010.
@article{967442, abstract = {{Several recent studies have explored the virtues of behavioral targeting and personalization for online advertising. In this paper, we add to this literature by proposing a cost-effective methodology for the prediction of demographic website visitor profiles that can be used for web advertising targeting purposes. The methodology involves the transformation of website visitors' clickstream patterns to a set of features and the training of Random Forest classifiers that generate predictions for gender, age, level of education and occupation category. These demographic predictions can support online advertisement targeting (i) as an additional input in personalized advertising or behavioral targeting, or (ii) as an input for aggregated demographic website visitor profiles that support marketing managers in selecting websites and achieving an optimal correspondence between target groups and website audience composition. The proposed methodology is validated using data from a Belgian web metrics company. The results reveal that Random Forests demonstrate superior classification performance over a set of benchmark algorithms. Further, the ability of the model set to generate representative demographic website audience profiles is assessed. The stability of the models over time is demonstrated using out-of-period data.}}, author = {{De Bock, Koen and Van den Poel, Dirk}}, issn = {{0169-2968}}, journal = {{FUNDAMENTA INFORMATICAE}}, keywords = {{out-of-period validation,Random Forests,ensemble classification,clickstream analysis,web advertising,web user profiling,demographic targeting,demographic prediction,AREA,SITES,EXPOSURE,BEHAVIOR,SELECTION,CLASSIFICATION,ONLINE,ROC CURVE,RANDOM FORESTS,BANNER ADVERTISEMENTS}}, language = {{eng}}, number = {{1}}, pages = {{49--70}}, title = {{Predicting website audience demographics for web advertising targeting using multi-website clickstream data}}, url = {{http://doi.org/10.3233/FI-2010-216}}, volume = {{98}}, year = {{2010}}, }
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: