Case-Based Reasoning

Davitkov Miroslav, 2011/3116

University of Belgrade
Faculty of Electrical Engineering
1. Case-Based Reasoning definition
Case-Based reasoning (CBR), broadly construed,
is the process of solving new problems
based on the solutions of similar past problems.
CBR is reasoning by remembering:
It is a starting point for new reasoning
Case-Based Reasoning is a well established research field
that involves the investigation of theoretical foundations,
system development and practical application building of
experience-based problem solving.
2 / 25
Anauto mechanic who fixes anengineby recallinganother car that
exhibitedsimilar symptoms
A lawyer who advocates a particular outcome in a trial based on
legal precedentsor ajudgewhocreatescaselaw.
An engineer copying working elements of nature (practicing
biomimicry), is treating nature as a database of solutions to
problems.
Case-basedreasoningisaprominent kindof analogymaking.
3 / 25
1. Case-Based Reasoning definition
Everyday examples of CBR :
1. Case previously made and stored experience item
4 / 25
2. CBR problem solver
2. Case-Base core of every case based problem solver
- collection of cases
One of the core assumptions behind CBR is that
similar problems have similar solutions.
A case-based problemsolver solves new problems primarily
byreuseof solutionsfromthecasesinthecase-base.
For thispurpose, oneor several relevant casesareselected.
5 / 25
2. CBR problem solver
6 / 25
Once similar cases are selected,
the solution(s) from the case(s) are adapted
to become a solution of the current problem.
2. CBR problem solver
When a new (successful) solution to the new problem is
found,
a new experience is made,
which can be stored in the case-base to increase its
competence,
thus implementing a learning behavior.
7 / 25
1. Structural (a common structured vocabulary, i.e. an ontology)
2. Textual (cases are represented as free text, i.e. strings)
3. Conversational
(a case is represented through a list of questions that varies from one
case to another ; knowledge is contained in customer / agent
conversations)
3. Types of CBR
There are three main types of CBR that differ significantly
from one another concerning case representation and
reasoning:
8 / 25
4. CBR Cycle
Despite the many different appearances of CBR
systems,
the essentials of CBR are captured in a surprisingly
simpleanduniformprocessmodel.
TheCBR cycleconsistsof 4sequential stepsaroundthe
knowledge of the CBR system.
TheCBR cycleisproposedbyAamodt andPlaza.
9 / 25
4. CBR Cycle
New Case
Retrieved Case
New Case
Solved Case
Tested /
Repaired
Case
Learned
Case
General Knowledge
Previous
Cases
Problem
Suggested
Solution
Confirmed
Solution
RETRIEVE
REUSE
REVISE
RETAIN
10 / 25
One or several cases from the case base are selected,
basedonthemodeledsimilarity.
Theretrieval task isdefinedasfindingasmall number of casesfrom
thecase-basewiththehighest similaritytothequery.
This is a k-nearest-neighbor retrieval task considering a specific
similarityfunction.
When the case base grows, the efficiency of retrieval decreases =>
methods that improve retrieval efficiency,
e.g. specific index structures such as kd-trees, case-retrieval nets, or
discrimination networks.
4.1. Retrieve
4. CBR Cycle
11 / 25
Reusing a retrieved solution can be quite simple if the solution is
returnedunchangedastheproposedsolutionfor thenewproblem.
Adaptation (if required, e.g. for synthetic tasks).
Several techniques for adaptation in CBR
- Transformational adaptation
- Generative adaptation
Most practical CBR applications today try to avoid extensive
adaptationfor pragmaticreasons.
4.2. Reuse
4. CBR Cycle
12 / 25
In this phase, feedback related to the solution constructed so far is
obtained.
Thisfeedback canbegivenintheformof acorrectnessratingof the
result or intheformof amanuallycorrected revised case.
The revised case or any other formof feedback enters the CBR
systemfor itsuseinthesubsequent retainphase.
4.3. Revise
4. CBR Cycle
13 / 25
The retain phase is the learning phase of a CBR system(adding a
revisedcasetothecasebase).
Explicit competence models have been developed that enable the
selectiveretentionof cases(becauseof thecontinuousincreaseof the
case-base).
The revised case or any other formof feedback enters the CBR
systemfor itsuseinthesubsequent retainphase.
4.4. Retain
4. CBR Cycle
14 / 25
5. CBR and the Future Internet
The development of the future internet is affected by two major
factors: semantics and collaboration.
Twoof themost influencingdevelopmentsof theSemanticWebare:
- theresourcedescriptionlanguageRDF (Resource Description
Framework)
- theknowledgerepresentationlanguageOWL (Web Ontology
Language), whichisbasedonRDF
Already beforethedevelopment of RDF and OWL, XML has been
used as a case representation within the case-based reasoning
community.
15 / 25
5. CBR and the Future Internet
There is a notable similarity between the ontologies developed
within semantic applications and the representation of cases in
structural case-basedreasoning.
Due to this similarity RDF and OWL both lend themselves to be
used as case representation languages and thus expand the
possibilitiesof case-basedreasoningwithinthegeneral WWW.
There are technological and methodological similarities between
ontologies and structured case-based reasoning and there are
synergiesthat canbereachedbymergingbothapproaches.
16 / 25
CaseML - an RDF based Case Markup Language (by Chen and Wu);
CaseML offers a domain-independent case ontology and also aims to
make case-based reasoning available within the Semantic Web.
SERVOGrid (by Aktas et al.) also uses RDF for case
representation;
It is embedded in a conversational case-based reasoning system
that aids scientists in finding resources such as program code or data
that are needed to solve a specific task
by assisting them in describing the necessary resources using meta
data.
5. CBR and the Future Internet
17 / 25
jCOLIBRI framework - OWL is being used as thecaseinterchange
language;
It is planned to advance the already distributed framework towards
an architecture consisting of Semantic Web Services (SWS)
whereproblemsolvingmethodsarerepresentedasWebServices;
In order to use these services the whole case-based reasoning process
is decomposed into single tasks,
which are then carried out by according Web Services.
5. CBR and the Future Internet
18 / 25
There is a close relation between collaborative filtering and CBR
and these two can benefit from each other.
Example 1: Collaborative filtering is used to assess the similarity
betweensongsinaCBR systemcreatingcustommusic compilations
(CoCoA) [Aguzzoli et al.].
Example2: A community based web search that uses theresults of
previous web searches of similar users in order to improve web
searchresults[BriggsandSmyth].
5. CBR and collaborative filtering
19 / 25
During the past twenty years, many CBR applications have been
developed, ranging fromprototypical applications build in research
labs to large-scale fielded applications developed by commercial
companies.
Application areas of CBR include:
- help-desk and customer service
- recommender systems in electronic commerce
- knowledge and experience management
- medical applications and applications in image processing
- applications in law, technical diagnosis, design, planning
- applications in the computer games and music domain.
6. CBR applications
20 / 25
We will compare CBR with the rule induction algorithm
of machine learning.
Like a rule-induction algorithm,
CBR starts with a set of cases or training examples;
it forms generalizations of these examples, albeit implicit ones,
by identifying commonalities between
a retrieved case and the target problem.
7. CBR compared to other methods
21 / 25
Thekey difference, however, between theimplicit generalization in
CBR and the generalization in rule induction lies in when the
generalization is made.
A rule-induction algorithmdraws its generalizations froma set of
trainingexamplesbeforethetarget problemisevenknown; that is, it
performseager generalization.
This is incontrast to CBR, whichdelays (implicit) generalizationof
itscasesuntil testingtime astrategyof lazy generalization.
CBR thereforetends to beagood approach for rich, complex
domainsinwhichtherearemyriadwaystogeneralizeacase.
7. CBR compared to other methods
22 / 25
Critics of CBR argue that it is an approach that accepts anecdotal
evidence as its main operating principle.
Without statistically relevant data for backing and implicit
generalization, there is no guarantee that the generalization is
correct.
There is recent work that develops CBR within a statistical
framework and formalizes case-based inference as a specific type of
probabilistic inference;
thus, it becomes possible to produce case-based predictions equipped
with a certain level of confidence.
8. Criticism of the CBR
23 / 25
The number of CBR approaches and applications developed up to
nowhasbecomequitelarge.
There is a significant number of CBR research groups and
commercial companies, which develop CBR methods, software
components, andapplicationsonaregular basis.
CBR is not only a technology but also a (process oriented)
method.
The combination of CBR with various other technologies within a
great bandwidth of applications has become increasingly attractive
for researchersaswell asbusinessprofessionals.
9. Conclusion
24 / 25
Ralph Bergmann, Klaus-Dieter Althoff, Mirjam
Minor, Meike Reichle, Kerstin Bach:
Case-Based Reasoning: Introduction and Recent
Developments
Benjamin Heitmann, Conor Hayes:
Enabling Case-Based Reasoning on the Web of Data
A. Aamodt, E. Plaza:
Case-Based Reasoning: Foundational Issues, Methodological
Variations, and System Approaches
10. References
Thank you for your attention!
Questions?
davitkov.miroslav@gmail.com
dm113116m@student.etf.rs