Mario Lenz

Case-Based Reasoning Technology

Extending some Concepts of CBR - Foundations of Case Retrieval Nets.- Diagnosis and Decision Support.- Intelligent Sales Support with CBR.- Textual CBR.- Using Configuration Techniques for Adaptation.- CBR Applied to Planning.- CBR for Design.- CBR for Experimental Software Engineering.- CBR for Tutoring and Help Systems.- CBR in Medicine.- Methodology for Building CBR Applications.- Related Areas.

Case Retrieval Nets: Basic Ideas and Extensions

An efficient retrieval of a relatively small number of relevant cases from a huge case base is a crucial subtask of Case-Based Reasoning. In this article, we present Case Retrieval Nets (CRNs), a memory model that has recently been developed for this task. The main idea is to apply a spreading activation process to a net-like case memory in order to retrieve cases being similar to a posed query case. We summarize the basic ideas of CRNs, suggest some useful extensions, and present some initial experimental results which suggest that CRNs can successfully handle case bases larger than considered usually in the CBR community.

Intelligent Sales Support with CBR

Electronic commerce applications are just about to leave their infancy. While electronic cash has been around for quite a few years, the amount of business carried out through the Internet is still relatively small compared to the potential of this young technology. There are plenty of reasons for this. Tennenbaum summarizes the barriers for using this medium today with the three words: confidence, convenience, and content. Customers must have confidence that their transactions are secure, their privacy is maintained, and they will not be subject to liability. It must be convenient, as simple to use as ATMs and as ubiquitous. Finally, there must be incentives to purchase goods via the Internet, be they a better price, service, or selection.

Question Answering with Textual CBR

In this paper, we show how case-based reasoning (CBR) techniques can be applied to document retrieval. The fundamental idea is to automatically convert textual documents into appropriate case representations and use these to retrieve relevant documents in a problem situation. In contrast to Information Retrieval techniques, we assume that a Textual CBR system focuses on a particular domain and thus can employ knowledge from that domain. We give an overview over our approach to Textual CBR, describe a particular application project, and evaluate the performance of the system.

Diagnosis and Decision Support

In this chapter, we will focus on the utilization of Case-Based Reasoning for solving problems in the area of diagnosis and decision support. For this, we will first discuss different types of analytic problem solving, explain alternative approaches of coping with specific problems, and finally sketch a number of successful applications.

Case-Based Reasoning: Survey and Future Directions

This paper surveys the field of case-based reasoning (CBR) – both in science and in industrial applications. It starts with a short introduction to the essential ideas and concepts CBR is built upon. Then follows a bit of history that is interesting for understanding the development and the current state of the field. Its main part introduces and reviews the most important sub-fields of CBR: theoretical foundations, CBR for document retrieval, product selection, help-desk support, diagnosis, configuration, planning, and design. In the last part, we discuss why the field has developed rather well and will have a promising future, particularly in new areas like self-service and e-commerce applications in the world wide web.

Applying Case Retrieval Nets to Diagnostic Tasks in Technical Domains

This paper presents Objectdirected Case Retrieval Nets, a memory model developed for an application of Case-Based Reasoning to the task of technical diagnosis. The key idea is to store cases, i.e. observed symptoms and diagnoses, in a network and to enhance this network with an object model encoding knowledge about the devices in the application domain.

CBR for Document Retrieval: The FALLQ Project

This paper reports about a project on document retrieval in an industrial setting. The objective is to provide a tool that helps finding documents related to a given query, such as answers in Frequently Asked Questions databases. A CBR approach has been used to develop a running prototypical system which is currently under practical evaluation.

Defining Knowledge Layers for Textual Case-Based Reasoning

Textual CBR applications deal with problems that have traditionally been addressed by the Information Retrieval community, namely the handling of textual documents. Since CBR is an AI technique, the questions arise as to what kind of knowledge may enhance the system, where this knowledge comes from, and how it contributes to the performance of such a system. We will address these questions in this paper by showing how the various pieces of knowledge available in a specific domain can be utilized.

IMTAS: Intelligent Multimedia Travel Agent System

This paper introduces IMTAS, an architecture enhancing existing tourist information systems with modern AI and multimedia technologies. IMTAS allows for a more flexible access to up-to-date information and supports new means of distribution in the tourism market.