Sascha Schmitt

A Similarity-Based Approach to Attribute Selection in User-Adaptive Sales Dialogs

For dynamic sales dialogs in electronic commerce scenarios, approaches based on an information gain measure used for attribute selection have been suggested. These measures consider the distribution of attribute values in the case base and are focused on the reduction of dialog length. The implicit knowledge contained in the similarity measures is neglected. Another important aspect that has not been investigated either is the quality of the produced dialogs, i.e. if the retrieval result is appropriate to the customers demands. Our approach takes the more direct way to the target products by asking the attributes that induce the maximum change of similarity distribution amongst the candidate cases, thereby faster discriminating the case base in similar and dissimilar cases. Evaluations show that this approach produces dialogs that reach the expected retrieval result with fewer questions. In real world scenarios, it is possible that the customer cannot answer a question. To nevertheless reach satisfactory results, one has to balance between a high information gain and the probability that the question will not be answered. We use a Bayesian Network to estimate these probabilities.

simVar : A Similarity-Influenced Question Selection Criterion for e-Sales Dialogs

Even though AI technologies like CBR have proved their strengths for intelligent sales support in EC applications, on-line customers often encounter e-sales systems that are hard to use. Before a search process is started, they either have to answer many annoying or irrelevant questions or they are faced with technical jargon of manufacturers they are not able to understand. On-line customers want personalised advice and adequate product offerings. Gaining sufficient information from the customer but also providing her with information at the right place is the key. Resulting from this fact, an automated communication process is needed that simulates the sales dialog between customers and human sales persons. This article proposes a method for question selection in e-sales dialogs based on the variance of the CBR systems inherent similarities. The method uses a similarity-influenced measure to reduce the number of questions required to find satisfactory products. Additionally, it is shown how questions can be selected on the level of abstraction appropriate to the customers knowledge.

Intelligent Customer Support for Product Selection with Case-Based Reasoning

Current product-oriented database search facilities are widely used on the Internet but recognized as limited in capability for intelligent sales support. The vision of intelligent knowledgeable virtual sales agents is to incorporate more knowledge about products, customers, and the sales process into an electronic shop. This chapter describes a knowledge-based technology called case-based reasoning (CBR) and shows how it can be adapted and applied for developing intelligent virtual sales agents. To emphasize the advantages for our approach, we implemented several applications, some of which are in daily use.

Collecting Experience on the Systematic Development of CBR Applications Using the INRECA Methodology

This paper presents an overview of the INRECA methodology for building and maintaining CBR applications. This methodology supports the collection and reuse of experience on the systematic development of CBR applications. It is based on the experience factory and the software process modeling approach from software engineering. CBR development experience is documented using software process models and stored in different levels of generality in a three-layered experience base. Up to now, experience from 9 industrial projects enacted by all INRECA II partners has been collected.

Entropy-Based vs. Similarity-Influenced: Attribute Selection Methods for Dialogs Tested on Different Electronic Commerce Domains

Recent research activities in the field of attribute selection for carrying on dialogs with on-line customers have focused on entropy-based approaches that make use of information gain measures. These measures consider the distribution of attribute values in the case base and are focused on their ability to reduce dialog length. The implicit knowledge contained in the similarity measures is neglected. In previous work, we proposed the similarity-influenced selection method simVar, which selects the attributes that induce the maximum change in similarity distribution amongst the candidate cases, thereby partitioning the case base into similar and dissimilar cases. In this paper we present an evaluation of the selection methods using three domains with distinct characteristics. The comparison of the selection methods is based on the quality of the dialogs generated. Statistical analysis was used to support the evaluation results.

A Product Customization Module Based on Adaptation Operators for CBR Systems in E-Commerce Environments

Existing electronic shops based on CBR technology allow customers to search for adequate products by only specifying the attributes for the products in a fuzzy way. Unfortunately, most electronic shops do not further support customers after the retrieval step. However, especially configurable products could be customized at this stage. Based on the approach of interactive adaptation operators, we present a flexible system architecture for a customization module which can be easily integrated in electronic shops. Our approach of a modular adaptation concept is implemented and currently tested within the ESPRIT project WEBSELL.

Knowledge-Intensive Product Search and Customization in Electronic Commerce

A major requirement of today’s online shops is the availability of competent virtual sales agents that guide the customers through the vast space of available products, services and other opportunities. This function is mostly implemented by search agents that should help customers to find relevant product information. While these search functions are considered quite important by the online sellers, the quality of the retrieval results is miserable [4]. The key to enhancing search quality and, more generally, to approaching the vision of intelligent, knowledgeable virtual sales agents, is to incorporate more knowledge about products, customers and the sales process into the sales agent. The quality of service becomes the dominating factor for achieving customer satisfaction and a good customer relationship. As a consequence customer relationship management [9] and knowledge management [10, 3] have been recognized as core disciplines with strategic importance for successful future business. In the context of companies which communicate with heir customers and partners via electronic online media, this requires one to make the company knowledge available and visible through the virtual agents that are supposed to be the primary access points to the company. This chapter describes a knowledge-based technology and related applications for developing intelligent virtual sales agents.