Rainer Maximini

An investigation of generalized cases

In the CBR literature from the past 25 years there is a considerable amount of research work that makes use of cases that are subspaces of some representation space rather than points in it. For cases of that kind, different terms have been used such as generalized case, prototype, schema, script, or abstract case. Our analysis of selected publications yields that on the one hand the same term is used for different concepts and on the other hand different terms are used for more or less the same concepts. So our goal is to improve the conceptual clarity by proposing an integrated classification schema for cases. We then use this schema to describe semantically founded ways for similarity definition and computation, depending on the class membership of query and case.

Case-based support for collaborative business

This paper describes the development of the generic collaboration support architecture CAKE incorporating case-based reasoning (CBR). CAKE provides unified access to knowledge available within an organization, and CBR technology is used throughout the system to distribute this knowledge to agents as required. Adaptive workflows and collaboration patterns selected by a CBR process are introduced for explicitly describing collaboration among agents. In order to guide the technical design of the architecture, a systematic analysis of the requirements for collaboration support has been performed in various application domains.

Towards collaborative agent-based knowledge support for time-critical and business-critical processes

In this paper, we present the Collaborative Agent-based Knowledge Engine approach for supporting mobile workers performing time-critical and/or business-critical tasks within agile projects. By a combination of sophisticated knowledge management and a light-weight workflow model, this approach provides guidance and knowledge as required to perform the individual activities. Moreover, we discuss aspects for maintaining project history, as well as possibilities for integrating tools regarding computer-supported collaborative work already deployed in organizations.

MINLP Based Retrieval of Generalized Cases

The concept of generalized cases has been proven useful when searching for configurable and flexible products, for instance, reusable components in the area of electronic design automation. This paper addresses the similarity assessment and retrieval problem for case bases consisting of traditional and generalized cases. While approaches presented earlier were restricted to continuous domains, this paper addresses generalized cases defined over mixed, continuous and discrete, domains. It extends the view on the similarity assessment as a nonlinear optimization problem (NLP) towards a mixed integer nonlinear optimization problem (MINLP), which is an actual research topic in mathematical optimization. This is an important step because most real world applications require mixed domains for the case description. Furthermore, we introduce two optimization-based retrieval methods that operate on a previously created index structure, which restricts the retrieval response time significantly.

Supporting Electronic Design Reuse by Integrating Quality-Criteria into CBR-Based IP Selection

The growing complexity of todays electronic designs requires reusing existing design components, called Intellectual Properties (IPs). Experience management approaches can be used to support design reuse, particularly the process of selecting reusable IPs. For the IP selection, quality criteria concerning the IP code and the documentation must be considered in addition to functional requirements of the IP. We analyse IP quality criteria in detail and show different concepts for their integration into the retrieval process.

Explanation Service for Complex CBR Applications

Case-based Reasoning (CBR) is a mature technology for building knowledge-based systems that are capable to produce useful results even if no answer matches the query exactly. Often the result sets presented to users are ordered by means of similarity and utility. However, for complex applications with knowledge intensive domains we have discovered that results sets enriched by calculated similarity values for particular answers are not sufficient. Users have a demand for additional information and explanations making the proposed results more transparent. By presenting additional explanations to them, their confidence in the result set increases and possible deficiencies, e.g., in the weight model, can be revealed and corrected. This paper presents a realized explanation service that combines several existing and new explanation technologies into one system.

Collaborative agent-based knowledge support for empirical and knowledge-intense processes

Independent from specific application domains, similar requirements can be identified regarding information needs during daily work. For coping with generality on the one hand and domain specificity on the other hand the Collaborative Agent-based Knowledge Engine CAKE is currently developed that combines agent and workflow technology in an innovative way. Agent technology is used for integrating various services, whereas workflow technology is used for coordinating collaboration among agents.

A knowledge representation format for virtual IP marketplaces

For the design of Systems on Chip (SoC) it is essential to reuse previously developed and verified virtual components in order to meet nowadays requirements in reliability, correctness, and time-to-market. On the downside, deciding about reusing a third-party component in a design situation can consume substantial time and resources. This is especially true in situations where many potential candidates exist due to the large amount of functional, non-functional, and quality related aspects of each component. In order to facilitate the search for components in an internet-based market scenario, we have developed a retrieval system that utilizes structural CBR. The approach relies on XML descriptions of each component that constitute the case base. In this paper we present IPCHL, which was originally intended to be a simple schema for such descriptions. As a consequence of the growing demands for structured documentation, IPCHL has become a representation format for virtual component descriptions incorporating many of the CBR knowledge containers.

Intelligent IP retrieval driven by application requirements

For the design of Systems on Chip it is essential to reuse previously developed and verified components, so-called Intellectual Properties (IPs), in order to meet nowadays requirements in reliability, correctness, and time-to-market. On the downside, deciding about reusing a third-party component in a design situation can consume substantial time and resources. This is especially true in situations where many potential candidates exist due to the large amount of functional, non-functional, and quality related aspects of each component. In order to facilitate the search for IPs in large IP assets, e.g. virtual marketplaces, we have developed the IP Broker tool suite that facilitates the development of IP retrieval solutions by making use of Case-based Reasoning technology. It enables integrators to search for IPs driven by their specific requirements. Applications based on IP Broker can be easily integrated into already existing company sites or portal sites.

Applying generalized cases to retrieval and configuration of life insurance policies

When searching for the right life insurance product, one can either confide in her/his insurance broker or use internet portal sites providing interactive assistance for finding suitable insurance products based on questionnaires. While the first alternative is risky, the second one is cumbersome because of many successive query formulations required. Since a single product is normally a tradeoff between customer requirements, the possibility of providing its priorities could speed up the search significantly. In this paper, we present an approach, which takes priorities in account, while searching for and configuring of insurance products. It makes use of Structural Case-Based-Reasoning (SCBR) extended by the concept of generalized cases. Here, each insurance product is represented as a generalized case what consequently allows the application of similarity assessment and retrieval techniques to the insurance domain.