Intelligent E-marketing with Web Mining, Personalization, and User-Adpated Interfaces - Advances in Data Mining - page 37

Intelligent E-marketing with Web Mining,
Personalization, and User-Adpated Interfaces
Petra Perner and G. Fiss
Institute of Computer Vision and Applied Computer Sciences
Arno-Nitzsche-Str. 45, 04277 Leipzig
[e-mail] http://www.ibai-research.de
Abstract. For many people the special attraction of E-commerce is linked to
the idea of being able to choose and order products and services directly on-line
from home. However, this is only one aspect of the new on-line sales model. As
in real sales processes competent counselling, in accordance with the
customer's necessities, and also after-sales assistance by help of the web play
an important part for the customer faith. This requires precise knowledge of the
customer's preferences who, however, in general does not like lengthy
questioning and the use of other communication routes. Holders of E-shops
have thus to gather the consumer's desires and preferences from his interactions
and the data resulting from the sales process, which requires a profound data
analysis. In this paper we describe what kind of data can be acquired in an e-
shop and how these data can be used to improve advertisement, marketing and
selling. We describe what kind of data mining methods are necessary and how
they can be applied to the data.
Keywords. E-commerce, Data Mining, User Profiling, Clickstream analysis,
Recommendation, User-Adapted Interfaces
1 Introduction
For many people the special attraction of E-commerce is linked to the idea of being
able to choose and order products and services directly on-line from home. However,
this is only one aspect of the new on-line sales model. As in real sales processes
competent counselling, in accordance with the customer's necessities, and also after-
sales assistance by help of the web play an important part for the customer faith. This
requires precise knowledge of the customer's preferences who, however, in general
does not like lengthy questioning and the use of other communication routes. Holders
of E-shops have thus to gather the consumer's desires and preferences from his
interactions and the data resulting from the sales process, which requires a profound
data analysis. This knowledge has then to be converted into an intelligent and, if
possible, entertaining presentation of the information wanted by the customer, where
multimedia means of expression are used, and without overstraining or understraining
him. Only in this way can he be motivated to continue the dialogue. As the capacities,
preferences and interests of the customers vary considerably in this field of
application, intelligent user guidance is indispensable.