Predicting customer response to cross-market discounts using ensemble methods
The globalization of markets and the growing number of companies increased the competition between retail companies. This reality affects all sectores of retail, from clothing to grocery. New ways to keep and gain customers help the companies staying competitives. Diferent strategies like publicity...
| Autor principal: | |
|---|---|
| Formato: | masterThesis |
| Idioma: | eng |
| Publicado em: |
2015
|
| Assuntos: | |
| Texto completo: | https://hdl.handle.net/10216/89669 |
| País: | Portugal |
| Oai: | oai:repositorio-aberto.up.pt:10216/89669 |
| Resumo: | The globalization of markets and the growing number of companies increased the competition between retail companies. This reality affects all sectores of retail, from clothing to grocery. New ways to keep and gain customers help the companies staying competitives. Diferent strategies like publicity and discounts are examples of tecnics used by the companies. One of the most recent tecniques is called cross-market discounts. This strategy consists of offering linked discounts in unrelated markets that have the same target customers but are not in direct competition with each other. The purpose of this thesis is to build models that allow the company predict the response of customers to the cross-market discounts. The company in study is a grocery retailer with a partnership with a gas company. The models will be supported by several data mining techniques and to enhace the performance, ensemble methods will be used. The contribution of this thesis is the implementation of ensemble methods in order to improve the models that predict the response to the cross-market discounts. |
|---|