Knowledge-based recommender applications provide valuable support for online customers in the identification products and services on e-Commerce platforms. State-of-the-art recommender technologies provide a couple of mechanisms for improving the accessibility of product assortments for customers, e.g. in situations where no solution can be found for a given set of customer requirements, the recommender application calculates a set of repair actions (minimal changes to the given customer requirements) which can guarantee the identification of solutions. Further examples for such mechanisms are explanations (why does a certain recommendation fit to the wishes and needs of a customer) or product comparisons (what are the major advantages of a certain product compared to other products in the result set), etc.
However, none of the existing recommender approaches exploit an underlying holistic view on the recommender process, i.e. take into account properties of the online customer such as domain knowledge, trust, available time, receptivity, etc. These and other aspects of an online sales dialogs are investigated within COHAVE (Consumer Behavior & Decision Modeling for Recommender Systems) in order to derive an integrated view on recommender processes. In this context, we analyze existing theories about customer behavior in online sales situations – especially, we focus on the areas of social psychology and cogitive psychology.
The goals of COHAVE were the following:
- Development of a dialog component which accompanies the online user in a sales dialog by taking into account psychological theories of consumer buying behavior (e.g. minimize time efforts, increasing the interest related to the product domain, etc.).
- Implementation of a knowledge acquisition component which allows the explicit definition of strategies how recommenders should behave in certain online buying contexts.
- Evaluation of the implemented environment within the context of user studies.