ARTIFICIAL INTELLIGENCE CASE-BASED REASONING FOR RECOMMENDER SYSTEMS PROF THOMAS ROTH-BERGHOFER, UNIVERSITY OF WEST LONDON, UK
Artificial Intelligence – Case-based reasoning for recommender systems – Invited talk at Techsylvania 2015
Aug 12, 2015
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A R T I F I C I A L I N T E L L I G E N C E C A S E - B A S E D R E A S O N I N G F O R R E C O M M E N D E R S Y S T E M S
P R O F T H O M A S R O T H - B E R G H O F E R , U N I V E R S I T Y O F W E S T L O N D O N , U K
– S H A N TA N U N A R AY E N , C E O , A D O B E
“We as consumers expect relevant, interactive, personalised material wherever we are.”
– D AV I D W A D H W A N I , S E N I O R V I C E P R E S I D E N T, A D O B E
“The new frontier is experience.”
E X P E R I E N C E
S P E C I F I C K N O W L E D G E O B TA I N E D F R O M A S P E C I F I C P R O B L E M S O LV I N G C O N T E X T
E X P E R I E N C E
S P E C I F I C K N O W L E D G E O B TA I N E D F R O M A S P E C I F I C P R O B L E M S O LV I N G C O N T E X T
E X P E R I E N C E
© U.S. Army Corps of Engineers - CC BY 2.0 https://www.flickr.com/photos/usacehq/10019330364
S P E C I F I C K N O W L E D G E O B TA I N E D F R O M A S P E C I F I C P R O B L E M S O LV I N G C O N T E X T
E X P E R I E N C E
© U.S. Army Corps of Engineers - CC BY 2.0 https://www.flickr.com/photos/usacehq/10019330364
© Craig Breil - CC BY 2.0https://www.flickr.com/photos/umich-msis/6550246787
S P E C I F I C K N O W L E D G E O B TA I N E D F R O M A S P E C I F I C P R O B L E M S O LV I N G C O N T E X T
E X P E R I E N C E
© U.S. Army Corps of Engineers - CC BY 2.0 https://www.flickr.com/photos/usacehq/10019330364
© Craig Breil - CC BY 2.0https://www.flickr.com/photos/umich-msis/6550246787
https://www.flickr.com/photos/marceldouwedekker/© Christoph Weigel - CC BY-SA 2.0
E X A M P L E S
E X A M P L E S
• Printer repairman remembers a similar failure of printer in the past.
E X A M P L E S
• Printer repairman remembers a similar failure of printer in the past.
Problem: My printout has white streaks. I printed on paper. Cleaning the printer did not help.
Solution: Ink cartridge was low on toner. ➜ Replace ink cartridge.
E X A M P L E S
• Printer repairman remembers a similar failure of printer in the past.
• A doctor remembers the symptoms of a bacterial infection from another patient.
Problem: My printout has white streaks. I printed on paper. Cleaning the printer did not help.
Solution: Ink cartridge was low on toner. ➜ Replace ink cartridge.
E X A M P L E S
• Printer repairman remembers a similar failure of printer in the past.
• A doctor remembers the symptoms of a bacterial infection from another patient.
Problem: My printout has white streaks. I printed on paper. Cleaning the printer did not help.
Solution: Ink cartridge was low on toner. ➜ Replace ink cartridge.
Problem: Patient X has fever. Blood test shows high count of white blood cells. Patient X tells about holidays in central Africa.
Solution: Tropical infection possible ➜ Prescribe further blood tests.
E X A M P L E S
• Printer repairman remembers a similar failure of printer in the past.
• A doctor remembers the symptoms of a bacterial infection from another patient.
Problem: My printout has white streaks. I printed on paper. Cleaning the printer did not help.
Solution: Ink cartridge was low on toner. ➜ Replace ink cartridge.
Problem: Patient X has fever. Blood test shows high count of white blood cells. Patient X tells about holidays in central Africa.
Solution: Tropical infection possible ➜ Prescribe further blood tests.
[ ]Problem +
SolutionCase =
E X A M P L E S
• Printer repairman remembers a similar failure of printer in the past.
• A doctor remembers the symptoms of a bacterial infection from another patient.
Problem: My printout has white streaks. I printed on paper. Cleaning the printer did not help.
Solution: Ink cartridge was low on toner. ➜ Replace ink cartridge.
Problem: Patient X has fever. Blood test shows high count of white blood cells. Patient X tells about holidays in central Africa.
Solution: Tropical infection possible ➜ Prescribe further blood tests.[ ]Problem
+ Solution
Case =
– D AV I D B . L E A K E , I N D I A N A U N I V E R S I T Y, U S A
“CBR is reasoning by remembering.”
Case-based reasoning is a cognitive approach for modelling human problem solving behaviour.
– A G N A R A A M O D T, N T N U , N O R W AY, A N D E N R I C P L A Z A , I I I A , S PA I N
“CBR is an approach to problem solving and learning.”
Case-based reasoning is an engineering approach for developing and implementing
intelligent systems.
Old Problem
B A S I C I D E A• Retrieve relevant experience from the case base.
• Re-use retrieved experience in the context of the current problem.
• Store new experience in the case base – learn.
New Problem Old Problem
Old Solution Case
i
Case Base
…
Old Problem
B A S I C I D E A• Retrieve relevant experience from the case base.
• Re-use retrieved experience in the context of the current problem.
• Store new experience in the case base – learn.
New Problem Old Problem
Old Solution Case
i
Case Base
Similarity
…
Old Problem
B A S I C I D E A• Retrieve relevant experience from the case base.
• Re-use retrieved experience in the context of the current problem.
• Store new experience in the case base – learn.
New Problem Old Problem
Old Solution Case
i
Case Base
New Solution
Similarity
…
Reuse
New Case
Old Problem
B A S I C I D E A• Retrieve relevant experience from the case base.
• Re-use retrieved experience in the context of the current problem.
• Store new experience in the case base – learn.
New Problem Old Problem
Old Solution Case
i
Case Base
New Solution
Similarity
…
Reuse
New Case
Old Problem
B A S I C I D E A• Retrieve relevant experience from the case base.
• Re-use retrieved experience in the context of the current problem.
• Store new experience in the case base – learn.
New Problem Old Problem
Old Solution Case
i
Case Base
New Solution
Similarity
…
Learn
Reuse
New Case
Old Problem
B A S I C I D E A• Retrieve relevant experience from the case base.
• Re-use retrieved experience in the context of the current problem.
• Store new experience in the case base – learn.
New Problem Old Problem
Old Solution Case
i
Case Base
New Solution
Similarity
…
Learn
Reuse
Key aspect: similarity
New Case
Old Problem
B A S I C R E C O M M E N D E R S Y S T E M
New Problem Old Problem
Old Solution Case
i
Case Base
New Solution
Similarity-based
retrieval
…
Products/ServicesRequirements match
New Case
Old Problem
B A S I C R E C O M M E N D E R S Y S T E M
New Problem Old Problem
Old Solution Case
i
Case Base
New Solution
Similarity-based
retrieval
…
myCBRCBRT O O L F O R R A P I D P R O T O T Y P I N G O F C B R A P P L I C AT I O N S
myCBRCBRT O O L F O R R A P I D P R O T O T Y P I N G O F C B R A P P L I C AT I O N S
F O R E X A M P L E : R E C O M M E N D E R S Y S T E M S
R E C O M M E N D E R S Y S T E M E X A M P L E
R E C O M M E N D E R S Y S T E M E X A M P L E
Body
Model
Color
Manufacturer
Mileage
S I M I L A R I T Y K N O W L E D G E
Similarity tables
Similarity functions
S I M I L A R I T Y- B A S E D R E T R I E VA L
S I M I L A R I T Y- B A S E D R E T R I E VA L
S I M I L A R I T Y- B A S E D R E T R I E VA L
S I M I L A R I T Y- B A S E D R E T R I E VA L
S I M I L A R I T Y- B A S E D R E T R I E VA L
R E U S E O F A U D I O M I X I N G E X P E R I E N C E
© CC0 1.0 Universal
R E U S E O F A U D I O M I X I N G E X P E R I E N C E
Case
➜ WorkflowBefore After
© CC0 1.0 Universal
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
Generate attributes1
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
Generate attributes1
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
Generate attributes1
Set data types2
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
Generate attributes1
Set data types2
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
Generate attributes1
Boolean
Concept
Float
Integer
String
Symbol
+
Set data types2
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
Generate attributes1
...
Set data types2
...
C S V I M P O R T
Colhead1 Colhead2 Colhead3 …
R1C1 R1C2 R1C3 …
R2C1 R2C2 R2C3 …
… … … …
Generate attributes1
...
Set data types2 Generate instances3
TA K E H O M E M E S S A G E
TA K E H O M E M E S S A G E
• Case-based reasoning is the method of choice whenever you need to compare products and services.
TA K E H O M E M E S S A G E
• Case-based reasoning is the method of choice whenever you need to compare products and services.
• myCBR Workbench provides tools to express similarity knowledge and an environment to test your knowledge model.
TA K E H O M E M E S S A G E
• Case-based reasoning is the method of choice whenever you need to compare products and services.
• myCBR Workbench provides tools to express similarity knowledge and an environment to test your knowledge model.
• myCBR SDK lets you easily integrate similarity-based retrieval in your (mobile) applications.
A R T I F I C I A L I N T E L L I G E N C E C A S E - B A S E D R E A S O N I N G F O R R E C O M M E N D E R S Y S T E M S
P R O F T H O M A S R O T H - B E R G H O F E R , U N I V E R S I T Y O F W E S T L O N D O N , U K
Prof Dr Thomas Roth-Berghofer Head of Research Cluster Digital Communities University of West London, United Kingdom
T H O M A S . R O T H - B E R G H O F E R @ U W L . A C . U K
Related Documents
