Fuzzy Web ad selector based on Web usage mining

Internet and Web technologies are widely available, making it easier for companies to conduct business and transfer information to customers. Moreover, they speed up financial transactions efficiently, reducing the transaction costs of commercial activities that businesses would normally incur. So, Internet business has created a competitive environment, a successful company wanting to survive and gain a competitive advantage must provide an acceptable bundle of customized services that satisfy customers´ needs. Despite the Internet´s obvious benefits as a new communication medium its advertising gives the same advertising messages to all customers and so has suffered from poor responses. To raise a Web ad´s effectiveness, we propose a Web ad selector that personalizes advertising messages for customers based on their preferences and interests. The Web ad selection system divides Web site customers with similar preferences into several segments through Web usage mining. It uses fuzzy rules that express customer segments´ surfing patterns on the basis of expert advice, and recommends appropriate ads by fuzzy inference.