An Industrial Application of Data Mining Techniques to Enhance the Effectiveness of On-Line Advertising

Abstract

Nowadays, online behavioural targeting is one of the most popular and profitable business strategies on the display advertising. It is based on data analysis of web user behaviours with the usage of machine learning aiming to optimise web advertising. The objective of this paper is to identify consumers who have no previously observed an advert but are “possible prospects” more likely to purchase an advertisement’s product. By identifying prospect customers, online advertisers may be able to optimise campaign performance, maximise their revenue as well as deliver advertisements tailored to a variety of user interests. Our work presents various benchmark machine-learning algorithms and attribute pre-processing techniques in the context of behavioural targeting. The performance of the experiments is evaluated using the key performance metric which is the predicted conversion rate. Our experimental results indicate that the presented data mining framework can significantly identify prospect customers in the vast majority of cases. Our results seem promising, indicating that there is a need for further studies in the area of data mining in online display advertising.