The analysis of the massive data sets collected by scientific instruments demands automation as a pre-requisite to analysis. There is an urgent need to create an intermediate level at which scientists can operate effectively; isolating them from the massive sizes and harnessing human analysis capabilities to focus on tasks in which machines do not even remotely approach humans---namely, creative data analysis, theory and hypothesis formation, and drawing insights into underlying phenomena. We give an overview of the main issues in the exploitation of scientific datasets, present five case studies where KDD tools play important and enabling roles, and conclude with future challenges for data mining and KDD techniques in science data analysis. keywords: Applications in Science, Data Analysis, overview article, large databases, automated analysis, scietific data sets, scientific discovery. To appear: Proceedings of the Second International Conference on Knowledge Discovery and Data Mining...

An efficient retrieval of a relatively small number of relevant cases from a huge case base is a crucial subtask of Case-Based Reasoning (CBR). In this article, we present Case Retrieval Nets, a memory model that has recently been developed for this task. The main idea is to apply a spreading activation process to the case memory structured as a Case Retrieval Net in order to retrieve cases being sufficiently similar to a posed query case. This article summarizes the basic ideas of Case Retrieval Nets and suggests some useful extensions. 1 Introduction A major area of research in recent years has been the development of techniques allowing for an efficient and yet flexible retrieval of relevant cases. This research has led to a number of sophisticated techniques for this subtask, as for example indexing techniques ([8]); kd--trees ([12, 13]); the heuristic Fish--and--Sink approach ([9]); the Crash memory model ([1]); and Knowledge-directed Spreading Activation (KDSA, [14]). In an earl...

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 customer's 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.

In this thesis, a specific memory structure is presented that has been developed for the retrieval task in Case-Based Reasoning systems, namely Case Retrieval Nets (CRNs). This model borrows from associative memories in that it suggests to interpret case retrieval as a process of re-constructing a stored case rather than searching for it in the traditional sense. Tow major advantages of this model are efficiency and flexibility: Efficiency, on the one hand, is concerned with the ability to handle large case bases and still deliver retrieval results reasonably fast. In this thesis, a formal investigation of efficiency is included but the main focus is set on a more pragmatic view in the sense that retrieval should, in the ideal case, be fast enough such that for the users of a related system no delay will be noticeable...

Although case-based reasoning (CBR) was introduced as an alternative to rule-based reasoning (RBR), there is a growing interest in integrating it with other reasoning paradigms, including RBR. New hybrid approaches are being piloted to achieve new synergies and improve problem-solving capabilities. In our approach to integration, CBR is used to satisfy multiple numeric constraints, and RBR allows the performance of “what if ” analysis needed for creative design. The domain of our investigation is nutritional menu planning. The task of designing nutritious, yet appetizing, menus is one at which human experts consistently outperform computer systems. Tailoring a menu to the needs of an individual requires satisfaction of multiple numeric nutrition constraints plus personal preference goals and aesthetic criteria. We first constructed and evaluated independent CBR and RBR menu planning systems, then built a hybrid system incorporating the strengths of each system. The hybrid outperforms either single strategy system, designing superior menus, while synergistically providing functionality that neither single strategy system could provide. In this paper, we present our hybrid approach, which has applicability to other design tasks in which both physical constraints and aesthetic criteria must be met.

Abstract not found

Abstract. Mémoire proposes a general framework for reasoning from cases in biology and medicine. Part of this project is to propose a memory organization capable of handling large cases and case bases as occur in biomedical domains. This article presents the essential principles for an efficient memory organization based on pertinent work in information retrieval. Information retrieval systems have been able to scale up to terabyte of data taking advantage of large databases research to build Internet search engines. They search for pertinent documents to answer a query using term-based ranking and/or global ranking schemes. Similarly, CBR systems search for pertinent cases using a scoring function for ranking the cases. Mémoire proposes a memory organization based on inverted indexes, which may be powered by databases to search and rank efficiently through large case bases. It can be seen as a first step toward largescale CBR systems, and in addition provides a framework for tight cooperation between CBR and IR. 1

This article presents some experimental results obtained from applying the Case Retrieval Net approach to large case bases. The obtained results suggest that CRNs can successfully handle case bases larger than considered in other reports. 1 Introduction In recent years, a major focus of research in the field of Case-Based Reasoning (CBR) has been directed to efficient case retrieval techniques. This is particularly true for those projects dealing with the application of CBR to real-world problem solving (whereas research on cognitive aspects sometimes gives emphasis to other properties). An example application area is the design of diagnosis and decision support systems ([8]). Here it is crucial that the relevant cases can be rapidly selected from large case bases. To satisfy this requirement, a number of techniques have been proposed recently (see [5] for an overview), including indexing schemes ([11]), kd--trees ([13]), the Fish--and--Sink approach ([12]), the Crash memory model ([3...

INduction and REasoning from CAses was the title of two large European CBR projects funded by the European Commission from 1992 -- 1999. In total, the two projects (abbreviated INRECA and INRECA-II) have obtained an overall funding of 3 MEuro, which enabled the 5 project partners to perform 55 person years of research and development work. The projects made several significant contributions to CBR research and helped shaping the European CBR community. The projects initiated the rise of three SMEs that base their main business on CBR application development, employing together more than 100 people in 2001. This paper gives an overview of the main research results obtained in both projects and provides links to the most important publications of the consortium.

This paper addresses the representation and the processing of background knowledge required for case-based reasoning applications in the field of classification, diagnosis, and decision support with the INRECA-system. This work is part of the