Intelligent Online Marketing : Predicting Conversion Rate Of New Keywords

Abstract: This thesis looks at the problem of predicting conversion rate of keywords in Google Adwords where little or no data for the keyword is available. Several methods are investigated and tested on data belonging to three different real world clients. The methods try to predict the conversion rate only given the keyword text. All methods are compared, using two different evaluation methods, with results showing good potential. Finally further improvements are suggested that could have a big impact on the results.