Bain, W. M.: Case-Based Reasoning: A Computer Model of Subjective Assessment. Ph.D. Thesis, Yale University, 1986.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....well de ned and bounded. This information includes: a database of fty precedent cases, the legal knowledge relevant to sentencing, and a hierarchical tree that contains over three hundred legal concepts relevant to sentencing at various levels of abstraction . Bain s aforementioned system, Judge [9, 10, 11], is, inter alia, case based (we have already seen that it adopts [a] hybrid approach involving both rule based and case based systems [86] Judge is a cognitive model of judges decision making when sentencing (and indeed it was based on interviews with judges) yet, it didn t have the aim of ....

Bain, W. M.: Case-Based Reasoning: A Computer Model of Subjective Assessment. Ph.D. Thesis, Yale University, 1986.

....E#ect of problem order: The second question addressed by these experiments is the e#ect on the learning and performance of ROBBIE of the order in which goals are presented to it. Problem order is well known to strongly a#ect learning in general, and learning of CBR systems in particular #e.g. #Bain, 1986; Redmond, 1992##. The question of how to select a sequence of situations to present to a system has also been examined in reference to the training of connectionist networks #e.g. #Cottrell Tsung, 1989; Elman, 1991##. problem order to a#ect the introspective learning of ROBBIE by altering the ....

Bain, W. #1986#. Case-basedReasoning: A Computer Model of Subjective Assessment. Ph.D. thesis, Yale University. Computer Science DepartmentTechnical Report 470.

....examples focusing on case based reasoning can be found in Reisbeck and Schank (1989) These include: IPP (Lebowitz 1980) a system that reads and understand stories about terrorist activities. CYRUS (Kolodner 1984) understands and reasons about the diplomatic travels of Cyrus Vance. JUDGE (Bain 1986) uses heuristics to determine sentences for legal cases. As new cases are read JUDGE updates its heuristics to insure an overall consistency in sentencing. CHEF (Hammon 1989) generates new Chinese stirfry and souffl e recipes by adapting old ones. It does this in a way that satisfies several goals ....

Bain W. M. 1986, Case-based Reasoning: A Computer Model of Subjective Assessment. PhD thesis, Yale University.

.... current one and then recommend that the user applies the previous case s solution to the new problem [Sha91, KA88] This reasoning may be given additional depth by building in a verification step before applying the solution [SM91] or by including some simple adaptation to the retrieved solution [Bai86] Thus EBR systems can be developed to provide the required shallow reasoning To provide an EBR system with a reasoning capability over a broad domain will require careful consideration of the indexing terms and similarity metrics that are used to discriminate between cases. One difficulty that ....

W. M. Bain. Case-Based Reasoning: A Computer Model of Subjective Assessment. PhD thesis, Yale University, 1986.

....human memory proposed by Schank and colleagues (Schank, 1982; Schank and Abelson, 1977) that have formed the basis of much work in CBR. Other CMSs with similar designs to CYRUS include MEDIATOR (Simpson, 1985) CHEF (Hammond, 1989a) PERSUADER (Sycara, 1988) PARADYME (Kolodner, 1989) and JUDGE (Bain, 1986). 4.7.1 Modelling DataLex DataLex (Tyree et al. 1988) reasons about legal cases which may arise when a person finds some object buried on another person s land, and the original owner of the object cannot be traced. Cases are indexed by a list of legal concepts which apply in such cases. The ....

Bain, W. M. (1986). Case-Based Reasoning: A Computer Model of Subjective Assessment.

....order had a distinct and dramatic effect on overall performance, and that introspective re indexing causes improved performance even as quality of problem order declines. It is acknowledged that problem order can strongly affect learning in general, and learning of CBR systems in particular (e.g. [4, 18]) A sequence of goals which gradually increases in complexity and distance from the original cases in memory, and which covers the range of possible situations, should maximize the effectiveness of the casebased reasoner s learning. Guided by the principles outlined in the previous paragraph, we ....

W.M. Bain. Case-based Reasoning: A Computer Model of Subjective Assessment. PhD thesis, Yale University, 1986. Computer Science Department Technical Report 470.

....justify and explain an interpretation of a case. Cases are hand coded into structures that support this process. Sadly, most CBR systems require the manual input of cases; they are not automatically generated from text or other sources (e.g. episodes in Kolodner s CYRUS [Kol84] and Bain s JUDGE [Bai86] used a conceptual dependency representation for their textual input. A human makes the decisions about how to structure a case and what indexing structures to use. A case s input form is extracted from text or created from events, and information is put into either an intermediate or final ....

W. M. Bain. Case-Based Reasoning: A Computer Model of Subjective Assessment. PhD thesis, Yale University, New Haven, CT, 1986.

.... analogy is found in the existing systems ARIES (Carbonell, 1986) and PRODIGY ANALOGY (Carbonell and Veloso, 1988) Substitution methods choose and install a replacement for some part of an old solution that does not fit the current situation requirements, as in CHEF (Hammond, 1986) JUDGE (Bain, 1989), CLAVIER (Hennessy and Hinkle, 1992) and MEDIATOR (Kolodner and Simpson, 1989) Transformation methods use heuristics to replace, delete, or add components to an old solution in order to make the old solution work in the new situation, as in CASEY (Koton, 1988) and JULIA (Hinrichs, 1992) The ....

Bain, W. (1989) Case-based Reasoning: A computer Model of Subjective Assessment. Ph.D. thesis, Yale University.

....Effect of problem order: The second question addressed by these experiments is the effect on the learning and performance of ROBBIE of the order in which goals are presented to it. Problem order is well known to strongly affect learning in general, and learning of CBR systems in particular (e.g. (Bain, 1986; Redmond, 1992) The question of how to select a sequence of situations to present to a system has also been examined in reference to the training of connectionist networks (e.g. Cottrell Tsung, 1989; Elman, 1991) problem order to affect the introspective learning of ROBBIE by altering the ....

Bain, W. (1986). Case-based Reasoning: A Computer Model of Subjective Assessment.

....uses the former to argue in support of the claim and the latter to make counter arguments. The result is a set of three ply arguments: arguments supporting a proposed solution, responses opposing those arguments, and a rebuttal. Many other works in interpretative CBR are also in the legal domain (Bain 1986, Branting 1988) The latest field where CBR seems to be also useful is creativity (Turner 1993, Kolodner, 1993b) The main working hypothesis is that much creativity stems from using old solutions in novel ways or combining old solutions in a different way. A creative artificial system would need ....

Bain W. (1986). Case-based reasoning: A computer model of subjective assessment. Ph.D.

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