Presentation on theme: "Case Based Reasoning Lecture 1: Introduction Professor Susan Craw"— Presentation transcript:
1 Case Based Reasoning Lecture 1: Introduction Professor Susan Craw B18a (via secretary)Lecture/Lab Notes available on the Virtual Campus and
2 Outline The Limitations of Rules Solving Problems Case Based Reasoning ApplicationsReading
3 The Limitations of Rules The success of rule-based expert systems is due to several factors:They can mimic some human problem-solving strategiesRules are a part of everyday life, so people can relate to themHowever, a significant limitation is the knowledge elicitation bottleneckExperts may be unable to articulate their expertiseHeuristic knowledge is particularly difficultExperts may be too busy…
4 Another Way We Solve Problems? By remembering how we solved a similar problem in the pastThis is Case Based Reasoning (CBR)memory-based problem-solvingre-using past experiencesExperts often find it easier to relate stories about past cases than to formulate rules
5 Elephants Never Forget! Some biologists suggest that elephants’ success in harsh environments may be due to their memories.A herd of elephants retains a collective memory of problems and their solutions:E.g., they remember where water can usually be found during a drought.Elephants can solve problems without using models or rules.
6 DatabasesDatabase technology would seem ideally suited to the task of retrieving known solutions to problemsDatabases are excellent at finding exact matches…But are poor at near or fuzzy matchesI’ve got the Answer What’s the Question?
7 The CBR Cycle Solution Review Retain Adapt Database Retrieve Similar NewProblem
8 R4 CycleRetrieve the cases from the case-base whose problem is most similar to the new problem.Reuse the solutions from the retrieved cases to create a proposed solution for the new problem.Revise the proposed solution to take account of the problem differences between the new problem and the problems in the retrieved cases.Retain the new problem and its revised solution as a new case for the case-base if appropriate.
9 Definitions of CBRCase-based reasoning is […] reasoning by rememberingA case-based reasoner solves new problems by adapting solutions that were used to solve old problemsCase-based reasoning is a recent approach to problem solving and learning […]Leake, 1996Riesbeck & Schank, 1989Aamodt & Plaza, 1994
10 CBR Assumption(s) The main assumption is that: Two other assumptions: Similar problems have similar solutions:e.g., an aspirin can be taken for any mild painTwo other assumptions:The world is a regular place: what holds true today will probably hold true tomorrow(e.g., if you have a headache, you take aspirin, because it has always helped)Situations repeat: if they do not, there is no point in remembering them(e.g., it helps to remember how you found a parking space near that restaurant)
11 Problems We Solve This Way Medicinedoctor remembers previous patients, especially for rare combinations of symptomsLawEnglish/US law depends on precedencecase histories are consultedManagementdecisions are often based on past rulingsFinancialperformance is predicted by past results
12 Good / Bad Applications for CBR Classification tasks (good for CBR)Diagnosis - what type of fault is this?Prediction / estimation - what happened when we saw this pattern before?Synthesis tasks (harder for CBR)Engineering DesignPlanningScheduling
13 Success Stories for CBR Failure predictionultrasonic NDT of rails for Dutch railwayswater in oil wells for SchlumbergerFailure analysisMercedes cars for DaimlerChryslersemiconductors at National SemiconductorMaintenance schedulingBoeing 737 enginesTGV trains for SNCFPlanningmission planning for US navyroute planning for DaimlerChrysler cars
14 Success Stories for CBR e-Commercesales support for standard productssales support for customised productsPersonalisationTV listings from Changing Worldsmusic on demand from Kirch Medianews stories via car radios for DaimlerBenzRe-Designgas taps for CopreciFormulation (recipes)rubber for racing tyres for Pirellicolouring plastics for General Electrictablets for AstraZeneca
15 Impact on Business @ Microsoft Within 9 months of introducing a CBR Microsoft’s call centre in GlasgowMicrosoft reported:10% increase in customer satisfaction rating28% increase in “first-time-fix” success rate13% increase in the “agent is informed” customer survey scoreA significant reduction in the time required to train new agentsMore consistent responses delivered by agents, regardless of the problem
16 CBR Honours Project Ideas CBR for filtering (anti-SPAM)Michael Long, BSc(Hons) 2004, SPAM filteringAmandine Orecchioni, 2005, ManagementCBR for DiagnosisKatya Ponce do Leon, MSc 2005, Fish Diagnosis for Marine LabGrant Gauld, BSc(Hons) 2005, CBR Helpdesk for Chevron-TexacoCBR for PlanningAbhishek Chakraborty, MSc 2005, CBR Healthcare Planning for Partners Research Emergency NutritionScott Morrice, BSc(Hons) 2004, “Killer Bunnies” gameIf you are interested in a CBR project next yearsee me or Nirmalie Wiratunga
17 Reading Article Books www.aaai.org/Library/Magazine/Vol26/vol26.html Tenth anniversary of the plastics color formulation tool, William Cheetham, AI Magazine, Vol 26, Fall, 2005.BooksI. Watson. Applying Knowledge Management: Techniques For Building Corporate Memories. Morgan Kaufmann, 2003.I. Watson. Applying Case-Based Reasoning: Techniques for Enterprise Systems. Morgan Kaufmann, 1997.
18 CBR Resources CBR on the web CBR Commercial Solutions CBR Commercial SolutionsOrenge fromKaidara Adviser from (www.kaidara.com)eGain (www.egain.com)Customer Service & Contact Centre SoftwareCBR Tools in our SchoolCBR-Works fromReCall fromWeka from