Aamodt, A. and Plaza, E. (1994) `Case-based reasoning: foundational issues, methodological variations and system approaches', Artificial Intelligence Communications, J., IOS Press, Vol. 7, No. 1, pp.39--59.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....humans knowledge since it is expressed via natural language. Consequently, in this work we use fuzzy sets theory to deal with linguistic values in the estimation by analogy procedure. 3. Estimation by analogy Estimation by analogy is essentially a form of CaseBased Reasoning which has four steps [1]: 1 Retrieve the most similar case or cases 2 Reuse the information and knowledge in that case to solve the problem 3 Revise the proposed solution 4 Retain the parts of this experience likely to be useful for future problem solving For effort estimation, CBR is based on the following ....

....for the proposition P i is closely similar project to P . Intuitively, P i is closely similar to P if d(P,P ) is in the vicinity of 1 (0 in the case of Euclidean distance) The only way to represent correctly the value vicinity of 1 is by using a fuzzy set defined in the unit interval [0, 1]. Indeed, this fuzzy set defines the closely similar qualification adopted in the environment. Figure 1 shows a possible representation for the value vicinity of 1 . In this example all projects that have d(P,P i ) higher than 0.5 contribute to the estimated cost of P; the contribution of each ....

A. Aamodt, E. Plaza, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", AI Communications, IOS Press, Vol. 7:1. 1994, pp. 39-59

....instances of these models (which might describe valuable experiences) become part of the organizational memory (OM) ready for reuse. Seeing the processes themselves as knowledge containers allows to describe their life and use cycle as a variant of the standard casebased reasoning (CBR) cycle ([4], 5] Figure 1 depicts this case oriented view: When a new business problem arises, an appropriate process model is retrieved and instantiated. The execution of the new instance embodies the reuse of the process knowledge as well as of the knowledge contained in the attached information needs. ....

Aamodt, A., and Plaza, E., "Case-based reasoning: Foundational issues, methodological variations, and system approaches", AI Communications, 7(1), March 1994.

....FIGURE 13. 10 FOLD CROSS VALIDATION TECHNIQUE [MLADENIC, 1996] 52 FIGURE 14. SYSTEM ARCHITECTURE. 58 FIGURE 15. CBR CYCLE [AAMODT AND PLAZA, 1994] . 61 FIGURE 16. AN EXAMPLE OF CASE REPRESENTATION IN THE RESTAURANTS DOMAIN . 63 FIGURE 17. RETRIEVE PHASE ....

....over the last years. Instead of relying just on general knowledge of a problem domain, or making associations along generalized relationships between problem descriptors and conclusions, CBR is able to utilize the specific knowledge of previously experienced, concrete problem situations (cases) [Aamodt and Plaza, 1994]. CBR is based on people reasoning. Humans are robust problem solvers; they routinely solve hard problems despite limited and uncertain knowledge, and their performance improves with experience. All of these qualities are desirable for real world AI systems. Consequently, it is natural to ask how ....

[Article contains additional citation context not shown here]

Aamodt, A., Plaza, E., "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches" .AICom - Artificial Intelligence Communications, IOS Press, Vol. 7: 1, pp. 39-59. 1994.

....base. It solves new problems by using solutions to old cases and adapting them to meet the requirements of the new problems [3] The process usually consists of four main activities: retrieve the similar cases, reuse the solutions, revise the solution, and retain the past experience as a new case [4]. Case retrieval is the first step and consists of two phases: search function for identifying a set of potential cases; and case matching, which is a more comprehensive similarity evaluation for ranking cases in the collected set [1] Case representation and case retrieval are critical for CBR, ....

Aamodt, A., and Plaza, E., `Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches', AI Communications, 7-1, 39-59, (1994).

....information for building and maintaining the USD student model. 2. Student modeling using Case Based Reasoning Techniques 2. 1 Introduction The main idea of CBR is to solve a new problem by retrieving a previous similar situation and by reusing information and knowledge of that situation [ 1 ]. Finding a similar past case and reusing its solution in the new problem situation solves a new problem. In CBR terminology, a case usually denotes a problem situation. A previously experienced situation, which has been captured and learned in a way that can be reused in the solving of future ....

A. Aamodt, E. Plaza, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", AI Communications, IOS Press, Vol. 7:1, pp. 39-59, 1994.

....through exception handling coordinator (EHC) EHC can at least record the exception for later use. SP EHC L3 Quality concern Figure 14 Close pattern of cross organizational exception handling 7. Intelligent Problem Solving Our exception handling system uses a case based reasoning (CBR) [22] based approach for managing the exception handling knowledge. This CBR mechanism is used to improve the exception handling capabilities. We have taken an approach of concept based case management [3] When an exception occurs and is propagated to the intelligent problem solver, the case ....

A. Aamodt, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", Artificial Intelligence Communications, IOS Press, Vol. 7: 1, 1994.

....in the NICU. The advice is context based. The results from the collaborations may affect the progress of the ongoing process coordination. The health professionals on board are not necessarily experienced in every aspect of intensive care. A case repository used in the case based reasoning (CBR) [5] based exception handling system stores valuable experience learned to help them make decisions. Exceptions are not avoidable in such an environment. An abnormal situation can cause special attention for healthcare professionals. Those abnormal situations should be resolved as soon as possible ....

....as exception handlers, along with other candidates such as ignore, retry, workflow recovery, and so on. Thus, several modification primitives are given in defeasible workflow that will ensure the modified workflows meet the correctness criteria established. 3. A case based reasoning (CBR) [5] based exception handling mechanism with integrated human involvements is used in defeasible workflow to support exception handling processes. This mechanism enhances the exception handling capabilities through collecting cases to capture experiences in handling experiences captured in those ....

[Article contains additional citation context not shown here]

A. Aamodt, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", Artificial Intelligence Communications, IOS Press, Vol. 7:1,1994

....knowledge. CBR is one of the knowledge based systems approaches the authors intend to apply in the domain. 5 CASE BASED REASONING CBR is an artificial intelligence (AI) approach to problem solving and learning that has received considerable attention over the last few years Kolodner (1993) Aamodt Plaza (1994). Applications of CBR include many domains, for example, meteorology, medicine, and telecommunications, Jones Roydhouse (1995) Schuster et al. 1997) Lewis (1993) CBR relies on the assumption that reminding and adaptation play a crucial role in human expert problem solving. Reminding means ....

Aamodt A. & Plaza E., 1994, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", AICOM, March, 7(1), pp. 39-59.

....was illustrated in [177] Although such discovery may be carried out automatically, the benefits of involvement of experts in new attribute selection are typically significant. 3. 5 Case based reasoning Case based reasoning (CBR) uses the knowledge of past experience when dealing with new cases [1, 100]. A case refers to a problem situation although, as with instance based learning [3] cases may be described with a simple attribute value vector, CBR most often uses a richer, often hierarchical data structure. CBR relies on a database of past cases that has to be designed in the way to ....

.... the new example [40] An optimal value of k may be determined automatically from the training set by using leaveone out cross validation [169] In our experiments in early diagnosis of rheumatic diseases [42] using the k NN algorithm implemented by Wettschereck [170] the best k from the range [1,75] was chosen in this manner. This implementation also incorporates feature weights determined from the training set. Namely, the contribution of each attribute to the distance 43 may be weighted, in order to avoid problems caused by irrelevant features [171] Let n = N at . Given two examples x = ....

[Article contains additional citation context not shown here]

Aamodt, A., Plaza, E., "Case-based reasoning: Foundational issues, methodological variations, and system approaches," AI Communications, 7(1) 39--59 (1994). 52

....is a recent approach to problem solving and learning that has gotten much attention over the last few years. 4. 1 Case Based Reasoning (CBR) Case Based Reasoning is used to solve new problems by remembering a previous similar situation and by reusing information and knowledge of that situation [1]. A case based reasoner generally has the following central tasks: identifying 7 the current problem situation, finding a past case similar to the new one, using that case to suggest a solution to the current problem, evaluating the proposed solution and updating the system by learning from this ....

Aamodt, A. and E. Plaza, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", 1994, AICOM, Vol.7, No.1, pp.39-59.

.... it involves matching the current problem against ones that have already been encountered in the past and reworking the solutions of the past problems in the current context. It can be represented as a cyclical process that is divided into the four following sub processes as depicted in Figure 1 (Aamodt Plaza 1994): retrieve the most similar case or cases from the case base . reuse the case to solve the problem . revise the proposed solution, if necessary . retain the solution for future problem solving New Case Retrieved Case Solved Case Tested Case Learned Case Previous Cases Retrieve ....

....the proposed solution, if necessary . retain the solution for future problem solving New Case Retrieved Case Solved Case Tested Case Learned Case Previous Cases Retrieve Reuse Revise Retain Problem Proposed Solution Confirmed Solution Figure 1 : The CBR cycle (adapted from Aamodt Plaza 1994) A new problem, described as a case, is compared to the existing cases in the case base and the most similar case or cases are retrieved. These cases are combined 6 and reused (i.e. adapted) to suggest a solution for the new problem. The solution proposed may need to be revised (i.e. evaluated ....

AAMODT A. & E. PLAZA 1994 "Case-based reasoning: foundational issues, methodological variations and system approaches", Artificial Intelligence Communications 7 (1) p.3059.

....rely on previous experience to guide problem solutions. Case based reasoning exploits this idea using AI systems designed to classify new cases and formulate solutions based on the evidence of specific cases already held in memory. A good introductory text to case based reasoning can be found in Aamodt Plaza (1994). In case based reasoning, a case usually refers to a problem situation. A case base (knowledge base) maintains previously experienced problems along with the corresponding solution in such a way that it can be reused in the solving of future problems. When a new situation is experienced, the ....

....the previous solution is applied to the new situation if the solution fails, the reason for that failure is identified and stored for future reference. Similarly, case based reasoning can be used for classification problems where the stored solution predicts the desired classification. Aamodt Plaza (1994) have described the case based reasoning method as a cycle described by four processes (and illustrated in figure 2.7 20 ) 1. Retrieve the most similar case(s) 2. Reuse case solution to solve problem. 20 Adapted from Aamodt Plaza (1994) 49 Case New Case Base Problem RETRIEVE REUSE ....

[Article contains additional citation context not shown here]

Aamodt, A. & Plaza, E. (1994), `Case-based reasoning: Foundational issues, methodological variations, and system approaches', Artificial Intelligence Communications 7(1), 39--59.

....knowledge contained in the cases. However, in many situations this specific knowledge by itself is not sufficient or appropriate to cope with all requirements of an application. Very often, background knowledge is available and or necessary to better explore and interpret the available cases [1]. Such general knowledge may state dependencies between certain case features and can be used to infer additional, previously unknown features from the known ones. Furthermore, some applications require an adaptation of a retrieved case according to the actual problem at hand [13, 6, 4] ....

....and the number of adults. This rule reliefs the user from the necessity to enter the value for the total number of persons when the number of children and the number of adults already have been entered. 2. 2 Adaptation Rules Adaptation rules come into play after a case has been retrieved [1]. Most likely, this case does not completely fit the requirements specified in the user s query: Some attributes of the retrieved case might differ somehow while others may exactly match the query. According to the differences, the retrieved case must be modified to become better suited to the ....

A. Aamodt and E. Plaza, `Case-based reasoning: Foundational issues, methodological variations, and system approaches ', AI Communications, 39--59, (1994).

....such as attribute value pairs and a specific type of similarity measure can usually be applied to compare different attribute values. 3 Textual CBR As mentioned before, the fundamental idea of CBR is to recall relevant cases from earlier problem solving episodes in a problem context [1]. In terms of a question answering system, this means that relevant documents should be recalled as answers to a question posed by some user. 3.1 The knowledge of a CBR system A crucial difference to the above sketched IR models is that CBR is a knowledgebased technique hence any CBR system ....

....AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA REUSE Retrieved Documents AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA AAAA Fig. 2. Process model of the Automatic Hotline based on the R 4 model [1] The Automatic Hotline directly utilizes a number of document types which are available at Siemens. These include usual FAQs, Informal Notes on recently observed problems as a simple verbal description of these problems and User Information Notes. While all three document types basically contain ....

Agnar Aamodt and Enric Plaza, `Case-based reasoning: foundational issues, methodological variations, and system approaches', AI Communications, 7(1), 39-- 59, (1994).

....introducing our methodology. II. The Case Based Diagnosis In the CBR terminology, a case denotes a (situation, solution) pair which expresses a problem previously solved, learned and stored in what we call a case memory. The process involved in CBR has been commonly described in (Kolodner,1993)(Aamodt Plaza, 1994) as a cyclical process including three steps : 1. Retrieve the most similar cases; 2. Adapt (reuse and revise) similar case solutions for matching it to the new situation; 3. Retain the new case. CBR tasks are divided into two classes, interpretative CBR and problem solving CBR (Leake, 1996) ....

....for any detailed entity there exists a corresponding abstraction. The case memory in (Roger and Karchenasse, 1994) uses non total abstractions : all detailed level entities are not necessarily mapped to the abstracted level. Source references of CBR (Kolodner, 1993) Riesbeck, Shank and Kass, 1994)(Aamodt and Plaza, 1994) (Bergmann, 1996) privilege the hierarchical organisation for the resolution of complex problems. Hierarchical memories accelerate the similarity assessment, they make the indexation task easier and in the case of entity abstraction, they permit to generate abstracted solutions. But the learning ....

, pp 39-52.

....similar problems have similar solutions, and that the types of problems encountered tend to recur [17] When these two observations hold true, it is worthwhile to solve new problems by reusing prior reasoning. The process by which a case based reasoner operates has been described by Aamodt Plaza [1] as a cyclical process comprised of the four REs: RETRIEVE the most similar case(s) REUSE the case(s) to solve the problem, REVISE the proposed solution if necessary, and RETAIN the new solution as a new case. The application of this CBR cycle to real problems raises a common set of issues, ....

Aamodt, A. and Plaza, E. "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", AI Communications 7(i), pp. 39-59.

....because the intellectual jump to multiple alignment is, perhaps, a little shorter than with those other languages. In this subsection, I first review how composite functions are executed in Prolog and then try to show how these concepts may translate into a framework of multiple alignment. [1] Prolog Fig. 4 shows how, in Prolog, the XOR function ( xor( may be combined with a negation function ( not( to make a negated XOR function ( notxor( Of course, the effect of notxor( could be achieved more simply by re writing the xor( function but it would not then illustrate ....

....meet the more recent goal of integrating learning with other kinds of computation. The aim in this section is to examine how aspects of language learning may be accommodated within the proposed broader frame of multiple alignment and unification, drawing on insights provided by the earlier work. [1] Learning segmental structure In terms of the SP model outlined in Section 4 of the companion article, learning may be seen as a process of comparing New information received from the environment with Old information already stored in memory, unifying the parts which match ( recognition ) or ....

[Article contains additional citation context not shown here]

) Aamodt, A., "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches," AI Communications, 7(1), pp. 39-59, 1994.

....at various levels of abstraction. In Paris, planning cases given at the concrete level are abstracted to several levels of abstraction which leads to a set of abstract cases that are stored in the case base. This case abstraction is done automatically in the retain phase of the CBR process modell [1]. When a new problem must be solved, an abstract case is retrieved whose problem description matches the current problem exactly at the abstract level. In the subsequent reuse phase, the abstract solution is refined, i.e. the details that are not contained in the abstract case are added to ....

A. Aamodt and E. Plaza, `Case-based reasoning: Foundational issues, methodological variations, and system approaches ', AI Communications, 7(1), 39--59, (1994).

....[Brod95] 2 MILESTONE is a collaborative project involving Trinity College Dublin, Broadcom ireann Research, Telecom ireann, and Ericssons which aims to provide a migration methodology supported by a generic tool 2. Case Based Reasoning for Legacy System Migration Case based reasoning, [Wats94, Kolo93, and Aamo96], has enjoyed widespread popularity as an alternative to expert systems for solution of problems of a certain type. CBR is useful in situations where a well understood model for the solution of a particular problem is unavailable. In these situations experts may be able to suggest solutions based ....

: Aamodt A. and Plaza E., "Case-Based Reasoning: foundational Issues, Methodological Variations, and System Approaches", Artifical Intelligence Communications, Vol. 7, No.1, 1996.

....can be various, and takes advantage of all the components, whether experimental or theoretical. It is strongly constrained by some specialized models in theoretical memory, called the points of view. 1. INTRODUCTION Case based reasoning is an AI methodology for processing experimental knowledge [2], a case being a set of empirical data. It has been originally proposed as an alternative to expert systems methodology [1] which is an AI methodology for processing theoretical knowledge. One way of defining theoretical knowledge is that it is knowledge that has lost all links with the objects ....

Aamodt Agnar and Plaza Enric, "Case-based reasoning : Foundational issues, methodological variations, and system approaches", AI Communications, 7(1), March 1994.

....database on economics. CBR is a popular artificial intelligence problem solving methodology that implements a form of computational analogy, where cases are commonly thought of as hproblem,solutioni pairs. The case based reasoning problem solving cycle can be defined by the following four steps (Aamodt and Plaza, 1994): ffl Retrieve: Given a new problem, retrieve a set of cases from a given case library whose problems are similar to the new problem. ffl Reuse: Attempt to apply the solutions of one or more of these stored cases to the new problem. ffl Revise: Based on feedback from the reuse step, adapt the ....

Aamodt, A. and E. Plaza (1994), "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches," AI Communications 7, 39--59.

....compared using a member of the family of L p norms. An L p norm is defined to be New Case Tested Repaired Case Learned Case Solved Case New Case Retrieved Case General Knowledge RETRIEVE REVISE REUSE RETAIN Problem Suggested Solution Confirmed Solution Previous Cases Figure 1. CBR Cycle (Aamodt and Plaza, 1994) Inducing Diagnostic Inference Models from Case Data 79 L x x p i i p i p ( x x 1 . The most common values for p are 1, 2, and and yield Manhattan distance, Euclidean distance, and max norm distance respectively. Specifically, these metrics can be computed as ....

Aamodt, A. and E. Plaza. 1994. "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches," AI Communications, Vol. 7, No. 1, pp. 39--59.

....A case is a set of empirical data. The general idea underlying CBR is that, in order to process a new case, it is preferable (more efficient, more exact, faster . to use one or several cases processed before. The classical CBR cycle, as presented on figure 2, which has been inspired by Aamodt [1] and Bichindaritz [2] follows five main steps: to abstract, to retrieve, to reuse, to revise and to retain. 2.1 Abstract The initial case representation is abstracted into an abstract case representation, in which the relevant features are calculated. The abstracted case is then matched with the ....

Aamodt Agnar and Plaza Enric, "Case-based reasoning : Foundational issues, methodological variations, and system approaches", AI Communications, 7(1), March 1994.

No context found.

Aamodt, A. and Plaza, E. (1994) `Case-based reasoning: foundational issues, methodological variations and system approaches', Artificial Intelligence Communications, J., IOS Press, Vol. 7, No. 1, pp.39--59.

No context found.

A. Aamodt, "Case-based reasoning: Foundational issues, methodological variations, and system approaches," Artificial Intelligence Communications, vol. 7, no. 1, 1994.

No context found.

A. Aamodt and E. Plaza, "Case-Based Reasoning: Foundational Issues, Methodological Variations and System Approaches", Artificial Intelligence Communications, 7(1), pp. 39-59, 1994.

No context found.

A.Aamodt, E.Plaza. "Case-Based Reasoning: Foundational Issues, Methodological Varations, and System Approaches". AICom-Artificial Intelligence Communications, IOS Press, Vol.7, pages 39-59.

No context found.

A. Aamodt and E. Plaza, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", AI Communications, 7, (1994) 39-59.

No context found.

A. Aamodt, E. Plaza, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", AI Communications, IOS Press, Vol. 7, no 1, 1994, pp. 39-59.

No context found.

A.Aamodt, E.Plaza. "Case-Based Reasoning: Foundational Issues, Methodological Varations, and System Approaches". AICom-Artificial Intelligence Communications, IOS Press, Vol.7, pages 39-59, 1994.

No context found.

Aamodt, A. & Plaza, E. (1994), "Case-Based Reasoning: Foundational Issues, Methodological Variations, and Systems Approaches", AI Communications, 7(1), pp. 39-59.

No context found.

Aamodt A., Plaza E.: `Case-Based Reasoning: Foundational Issues, methodological Variations, and System Approaches. AI Communications', IOS Press, Vol. 7: 1, pp. 39-59, (1994)

No context found.

Agnar Aamodt and Enric Plaza, "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approach," Artificial Intelligence Communications, Vol. 7, No. 1, 1994, pp. 39-59.

No context found.

Aamodt A., Plaza E., "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches", AICOM, March, 7(1), pp39-59.

No context found.

Aamodt, A., Plaza, E., "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches.", AICOM, vol. 7, no. 1, March 1994.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC