Measuring the Precision of Multi-perspective Process Models

Felix Mannhardt
Felix MannhardtResearch Scientist at SINTEF
β€’β€’
Measuring the Precision of
Multi-perspective
Process Models
Felix Mannhardt
joint work with
Massimiliano de Leoni, Hajo A. Reijers,
Wil M.P. van der Aalst
Precision
Department of Mathematics and Computer Science PAGE 1 / 8
β€œFlower Model” lacking any precision
B
C
A
Precision of Multi-perspective Process Models
Department of Mathematics and Computer Science PAGE 2 / 8
A π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐴
B
π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐡
π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐡 > π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐴
Existing work ignores added precision
by multi-perspective rules / constraints
Approach: Multi-perspective Precision
Department of Mathematics and Computer Science PAGE 3 / 8
Multi-perspective
Process Model (P)
Fitting Event
Log (L)
Precision
[0..1]
INPUT OUTPUT
APPROACH
𝒆 βˆˆπ‘³
𝒐𝒃𝒔𝒆𝒓𝒗𝒆𝒅 𝑷 (𝒆)
π’‘π’“π’†π’„π’Šπ’”π’Šπ’π’(𝑷, 𝑳) =
𝒆 βˆˆπ‘³
π’‘π’π’”π’”π’Šπ’ƒπ’π’† 𝑷(𝒆)
Precision: Observed / Possible Behavior
Department of Mathematics and Computer Science PAGE 4 / 8
C Id Event Loan obs pos
1 𝑒1 Handle Request 800
1 𝑒2 Simple Check -
1 𝑒3 Decide -
2 𝑒4 Handle Request 1800
2 𝑒5 Ext. Check -
2 𝑒6 Decide -
𝒐𝒃𝒔 𝑷 𝒆 𝟏 = { Handle Request } = 𝟏
𝒑𝒐𝒔 𝑷 𝒆 𝟏 = Handle Request = 𝟏
𝒔𝒕𝒂𝒕𝒆 𝒆 𝟏 = (< >, { })
C Id Event Loan obs pos
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check -
1 𝑒3 Decide -
2 𝑒4 Handle Request 1800
2 𝑒5 Ext. Check -
2 𝑒6 Decide -
C Id Event Loan obs pos
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 1
1 𝑒3 Decide -
2 𝑒4 Handle Request 1800
2 𝑒5 Ext. Check -
2 𝑒6 Decide -𝒐𝒃𝒔 𝑷 𝒆 𝟐 = { Simple Check } = 𝟏
𝒑𝒐𝒔 𝑷 𝒆 𝟐 = Simple Check = 𝟏
𝒔𝒕𝒂𝒕𝒆 𝒆 𝟐 = (< 𝐻 >, { 𝐿 = 800 })
C Id Event Loan obs pos
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 1
1 𝑒3 Decide - 1 1
2 𝑒4 Handle Request 1800
2 𝑒5 Ext. Check -
2 𝑒6 Decide -
𝒐𝒃𝒔 𝑷 𝒆 πŸ‘ = { Decide } = 𝟏
𝒑𝒐𝒔 𝑷 𝒆 πŸ‘ = Decide = 𝟏
𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ‘ = (< 𝐻, 𝑆 >, { 𝐿 = 800 })
C Id Event Loan obs pos
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 1
1 𝑒3 Decide - 1 1
2 𝑒4 Handle Request 1800 1 1
2 𝑒5 Ext. Check -
2 𝑒6 Decide -
𝒐𝒃𝒔 𝑷 𝒆 πŸ’ = { Handle Request } = 𝟏
𝒑𝒐𝒔 𝑷 𝒆 πŸ’ = Handle Request = 𝟏
𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ’ = (<>, { })
C Id Event Loan obs pos
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 1
1 𝑒3 Decide - 1 1
2 𝑒4 Handle Request 1800 1 1
2 𝑒5 Ext. Check - 1 2
2 𝑒6 Decide -𝒐𝒃𝒔 𝑷 𝒆 πŸ“ = { Ext. Check } = 𝟏
𝒑𝒐𝒔 𝑷 𝒆 πŸ“ = Ext. Ceπ‘π‘˜, π‘†π‘–π‘šπ‘π‘™π‘’ πΆβ„Žπ‘’π‘π‘˜ = 𝟐
𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ“ = (< 𝐻 >, { 𝐿 = 1800 })
C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 1
1 𝑒3 Decide - 1 1
2 𝑒4 Handle Request 1800 1 1
2 𝑒5 Ext. Check - 1 2
2 𝑒6 Decide - 1 1
𝒐𝒃𝒔 𝑷 𝒆 πŸ” = Decide = 𝟏
𝒑𝒐𝒔 𝑷 𝒆 πŸ” = 𝐷𝑒𝑐𝑖𝑑𝑒 = 𝟏
𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ” = (< 𝐻, 𝐸 >, { 𝐿 = 1800 })
C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 1
1 𝑒3 Decide - 1 1
2 𝑒4 Handle Request 1800 1 1
2 𝑒5 Ext. Check - 1 2
2 𝑒6 Decide - 1 1
6 7
𝒐𝒃𝒔 𝑷 𝒆 = { observed activities at state }
𝒑𝒐𝒔 𝑷 𝒆 = | π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ π‘Žπ‘π‘‘π‘–π‘£π‘–π‘‘π‘–π‘’π‘  π‘Žπ‘‘ π‘ π‘‘π‘Žπ‘‘π‘’ |
𝒔𝒕𝒂𝒕𝒆 𝒆 = π‘ π‘‘π‘Žπ‘‘π‘’ π‘œπ‘“ π‘‘β„Žπ‘’ π‘π‘Ÿπ‘œπ‘π‘’π‘ π‘  π‘šπ‘œπ‘‘π‘’π‘™
π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝑃, 𝐿 =
π‘’βˆˆπΏ π‘œπ‘π‘ π‘’π‘Ÿπ‘£π‘’π‘‘ 𝑃(𝑒) = 6
π‘’βˆˆπΏ π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ 𝑃(𝑒) = 7
β‰ˆ 𝟎. πŸ–πŸ”
Full Example for Model A & Model B
Department of Mathematics and Computer Science PAGE 5 / 8
A
C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 2
1 𝑒3 Decide - 1 1
2 𝑒4 Handle Request 1800 1 1
2 𝑒5 Extensive Check - 2 2
2 𝑒6 Decide - 1 1
3 𝑒7 Handle Request 1800 1 1
3 𝑒8 Simple Check - 2 2
3 𝑒9 Decide - 1 1
4 𝑒10 Handle Request 2500 1 1
4 𝑒11 Extensive Check - 1 2
4 𝑒12 Decide - 1 1
14 16
Model A
π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝑃, 𝐿 =
π‘’βˆˆπΏ π‘œπ‘π‘ π‘’π‘Ÿπ‘£π‘’π‘‘ 𝑃(𝑒) = 14
π‘’βˆˆπΏ π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ 𝑃(𝑒) = 16
β‰ˆ 𝟎. πŸ–πŸ•πŸ“
B
C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷
1 𝑒1 Handle Request 800 1 1
1 𝑒2 Simple Check - 1 1
1 𝑒3 Decide - 1 1
2 𝑒4 Handle Request 1800 1 1
2 𝑒5 Extensive Check - 2 2
2 𝑒6 Decide - 1 1
3 𝑒7 Handle Request 1800 1 1
3 𝑒8 Simple Check - 2 2
3 𝑒9 Decide - 1 1
4 𝑒10 Handle Request 2500 1 1
4 𝑒11 Extensive Check - 1 1
4 𝑒12 Decide - 1 1
14 14
Model B
π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝑃, 𝐿 =
π‘’βˆˆπΏ π‘œπ‘π‘ π‘’π‘Ÿπ‘£π‘’π‘‘ 𝑃(𝑒) = 14
π‘’βˆˆπΏ π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ 𝑃(𝑒) = 14
β‰ˆ 𝟏
Evaluation on Road Fines Log
0.30
0.36
0.64
0.83
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Inductive Miner Inductive Miner &
Rules
Normative Model Normative Model &
Rules
ETC Precision Precision Fitness
Summary
Department of Mathematics and Computer Science PAGE 7 / 8
β€’ 1st precision measure for
multi-perspective process models
βˆ’ Fast to calculate
βˆ’ Flexible framework
βˆ’ Implemented in ProM
β€’ Preliminary Evaluation
βˆ’ Illustrative examples
βˆ’ Real-life dataset with > 500,000 events
Handle
750
Simple Decide
Handle
1250
Ext. Decide
Handle
5000
Simple Decide
Handle
750
Simple Decide
Handle
1500
Simple Decide
precision [0..1]
Department of Mathematics and Computer Science
Image source: http://commons.wikimedia.org/wiki/File:Pictofigo_-_Idea.png
Questions? Remarks? Ideas?
1 of 9

Recommended

Aecor. Purely functional event sourcing by
Aecor. Purely functional event sourcingAecor. Purely functional event sourcing
Aecor. Purely functional event sourcingDenis Mikhaylov
811 viewsβ€’33 slides
JavaCro'14 - JCalc Calculations in Java with open source API – Davor Sauer by
JavaCro'14 - JCalc Calculations in Java with open source API – Davor SauerJavaCro'14 - JCalc Calculations in Java with open source API – Davor Sauer
JavaCro'14 - JCalc Calculations in Java with open source API – Davor SauerHUJAK - Hrvatska udruga Java korisnika / Croatian Java User Association
1.9K viewsβ€’13 slides
JCalc:Calculations in java with open source API by
JCalc:Calculations in java with open source APIJCalc:Calculations in java with open source API
JCalc:Calculations in java with open source APIDavor Sauer
766 viewsβ€’13 slides
ORM is an Offensive Anti-Pattern by
ORM is an Offensive Anti-PatternORM is an Offensive Anti-Pattern
ORM is an Offensive Anti-PatternYegor Bugayenko
1.4K viewsβ€’24 slides
Chapter 5 by
Chapter 5Chapter 5
Chapter 5EasyStudy3
23 viewsβ€’38 slides
Decision Mining Revisited - Discovering Overlapping Rules by
Decision Mining Revisited - Discovering Overlapping RulesDecision Mining Revisited - Discovering Overlapping Rules
Decision Mining Revisited - Discovering Overlapping RulesFelix Mannhardt
1.1K viewsβ€’22 slides

More Related Content

Similar to Measuring the Precision of Multi-perspective Process Models

CREW SCHEDULING by
CREW SCHEDULINGCREW SCHEDULING
CREW SCHEDULINGEvren E
1.1K viewsβ€’25 slides
Synco tm 700 controller operation and commissioning part 1 by
Synco tm 700 controller operation and commissioning part 1Synco tm 700 controller operation and commissioning part 1
Synco tm 700 controller operation and commissioning part 1CONTROLS & SYSTEMS
95 viewsβ€’22 slides
Flow system control by
Flow system controlFlow system control
Flow system controlSaif al-din ali
231 viewsβ€’15 slides
Hybrid rule engines (rulesfest 2010) by
Hybrid rule engines (rulesfest 2010)Hybrid rule engines (rulesfest 2010)
Hybrid rule engines (rulesfest 2010)Geoffrey De Smet
1.2K viewsβ€’59 slides
Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T... by
Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T...Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T...
Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T...Ahmed Gad
1.6K viewsβ€’97 slides
Unit 4--probability and probability distribution (1).pptx by
Unit 4--probability and probability distribution (1).pptxUnit 4--probability and probability distribution (1).pptx
Unit 4--probability and probability distribution (1).pptxakshay353895
34 viewsβ€’83 slides

Similar to Measuring the Precision of Multi-perspective Process Models(12)

CREW SCHEDULING by Evren E
CREW SCHEDULINGCREW SCHEDULING
CREW SCHEDULING
Evren Eβ€’1.1K views
Synco tm 700 controller operation and commissioning part 1 by CONTROLS & SYSTEMS
Synco tm 700 controller operation and commissioning part 1Synco tm 700 controller operation and commissioning part 1
Synco tm 700 controller operation and commissioning part 1
CONTROLS & SYSTEMSβ€’95 views
Hybrid rule engines (rulesfest 2010) by Geoffrey De Smet
Hybrid rule engines (rulesfest 2010)Hybrid rule engines (rulesfest 2010)
Hybrid rule engines (rulesfest 2010)
Geoffrey De Smetβ€’1.2K views
Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T... by Ahmed Gad
Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T...Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T...
Introduction to Artificial Neural Networks (ANNs) - Step-by-Step Training & T...
Ahmed Gadβ€’1.6K views
Unit 4--probability and probability distribution (1).pptx by akshay353895
Unit 4--probability and probability distribution (1).pptxUnit 4--probability and probability distribution (1).pptx
Unit 4--probability and probability distribution (1).pptx
akshay353895β€’34 views
Dti2143 chapter 3 arithmatic relation-logicalexpression by alish sha
Dti2143 chapter 3 arithmatic relation-logicalexpressionDti2143 chapter 3 arithmatic relation-logicalexpression
Dti2143 chapter 3 arithmatic relation-logicalexpression
alish shaβ€’516 views
How to Develop Your Own Simulators for Discrete-Event Systems by Donghun Kang
How to Develop Your Own Simulators for Discrete-Event SystemsHow to Develop Your Own Simulators for Discrete-Event Systems
How to Develop Your Own Simulators for Discrete-Event Systems
Donghun Kangβ€’456 views
12 speed gear box by Gopinath Guru
12 speed gear box12 speed gear box
12 speed gear box
Gopinath Guruβ€’21.7K views

More from Felix Mannhardt

A Taxonomy for Combining Activity Recognition and Process Discovery in Indust... by
A Taxonomy for Combining Activity Recognition and Process Discovery in Indust...A Taxonomy for Combining Activity Recognition and Process Discovery in Indust...
A Taxonomy for Combining Activity Recognition and Process Discovery in Indust...Felix Mannhardt
527 viewsβ€’20 slides
Estimating the Impact of Incidents on Process Delay - ICPM 2019 by
Estimating the Impact of Incidents on Process Delay - ICPM 2019Estimating the Impact of Incidents on Process Delay - ICPM 2019
Estimating the Impact of Incidents on Process Delay - ICPM 2019Felix Mannhardt
559 viewsβ€’24 slides
Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro... by
Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro...Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro...
Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro...Felix Mannhardt
1.1K viewsβ€’21 slides
Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo... by
Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo...Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo...
Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo...Felix Mannhardt
1.2K viewsβ€’11 slides
From Low-Level Events to Activities - A Pattern-based Approach by
From Low-Level Events to Activities - A Pattern-based ApproachFrom Low-Level Events to Activities - A Pattern-based Approach
From Low-Level Events to Activities - A Pattern-based ApproachFelix Mannhardt
143 viewsβ€’24 slides
Analyzing the Trajectories of Patients with Sepsis using Process Mining by
Analyzing the Trajectories of Patients with Sepsis using Process MiningAnalyzing the Trajectories of Patients with Sepsis using Process Mining
Analyzing the Trajectories of Patients with Sepsis using Process MiningFelix Mannhardt
1.2K viewsβ€’9 slides

More from Felix Mannhardt(7)

A Taxonomy for Combining Activity Recognition and Process Discovery in Indust... by Felix Mannhardt
A Taxonomy for Combining Activity Recognition and Process Discovery in Indust...A Taxonomy for Combining Activity Recognition and Process Discovery in Indust...
A Taxonomy for Combining Activity Recognition and Process Discovery in Indust...
Felix Mannhardtβ€’527 views
Estimating the Impact of Incidents on Process Delay - ICPM 2019 by Felix Mannhardt
Estimating the Impact of Incidents on Process Delay - ICPM 2019Estimating the Impact of Incidents on Process Delay - ICPM 2019
Estimating the Impact of Incidents on Process Delay - ICPM 2019
Felix Mannhardtβ€’559 views
Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro... by Felix Mannhardt
Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro...Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro...
Data-driven Process Discovery - Revealing Conditional Infrequent Behavior fro...
Felix Mannhardtβ€’1.1K views
Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo... by Felix Mannhardt
Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo...Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo...
Unsupervised Event Abstraction using Pattern Abstraction and Local Process Mo...
Felix Mannhardtβ€’1.2K views
From Low-Level Events to Activities - A Pattern-based Approach by Felix Mannhardt
From Low-Level Events to Activities - A Pattern-based ApproachFrom Low-Level Events to Activities - A Pattern-based Approach
From Low-Level Events to Activities - A Pattern-based Approach
Felix Mannhardtβ€’143 views
Analyzing the Trajectories of Patients with Sepsis using Process Mining by Felix Mannhardt
Analyzing the Trajectories of Patients with Sepsis using Process MiningAnalyzing the Trajectories of Patients with Sepsis using Process Mining
Analyzing the Trajectories of Patients with Sepsis using Process Mining
Felix Mannhardtβ€’1.2K views
XESLite - Handling Event Logs in ProM by Felix Mannhardt
XESLite - Handling Event Logs in ProMXESLite - Handling Event Logs in ProM
XESLite - Handling Event Logs in ProM
Felix Mannhardtβ€’1.4K views

Recently uploaded

scopus cited journals.pdf by
scopus cited journals.pdfscopus cited journals.pdf
scopus cited journals.pdfKSAravindSrivastava
15 viewsβ€’15 slides
Indian council for child welfare by
Indian council for child welfareIndian council for child welfare
Indian council for child welfareRenuWaghmare2
7 viewsβ€’21 slides
Note on the Riemann Hypothesis by
Note on the Riemann HypothesisNote on the Riemann Hypothesis
Note on the Riemann Hypothesisvegafrank2
8 viewsβ€’20 slides
Exploring the nature and synchronicity of early cluster formation in the Larg... by
Exploring the nature and synchronicity of early cluster formation in the Larg...Exploring the nature and synchronicity of early cluster formation in the Larg...
Exploring the nature and synchronicity of early cluster formation in the Larg...SΓ©rgio Sacani
1.4K viewsβ€’12 slides
Bacterial Reproduction.pdf by
Bacterial Reproduction.pdfBacterial Reproduction.pdf
Bacterial Reproduction.pdfNandadulalSannigrahi
34 viewsβ€’32 slides
Oral_Presentation_by_Fatma (2).pdf by
Oral_Presentation_by_Fatma (2).pdfOral_Presentation_by_Fatma (2).pdf
Oral_Presentation_by_Fatma (2).pdffatmaalmrzqi
8 viewsβ€’7 slides

Recently uploaded(20)

Indian council for child welfare by RenuWaghmare2
Indian council for child welfareIndian council for child welfare
Indian council for child welfare
RenuWaghmare2β€’7 views
Note on the Riemann Hypothesis by vegafrank2
Note on the Riemann HypothesisNote on the Riemann Hypothesis
Note on the Riemann Hypothesis
vegafrank2β€’8 views
Exploring the nature and synchronicity of early cluster formation in the Larg... by SΓ©rgio Sacani
Exploring the nature and synchronicity of early cluster formation in the Larg...Exploring the nature and synchronicity of early cluster formation in the Larg...
Exploring the nature and synchronicity of early cluster formation in the Larg...
SΓ©rgio Sacaniβ€’1.4K views
Oral_Presentation_by_Fatma (2).pdf by fatmaalmrzqi
Oral_Presentation_by_Fatma (2).pdfOral_Presentation_by_Fatma (2).pdf
Oral_Presentation_by_Fatma (2).pdf
fatmaalmrzqiβ€’8 views
ALGAL PRODUCTS.pptx by RASHMI M G
ALGAL PRODUCTS.pptxALGAL PRODUCTS.pptx
ALGAL PRODUCTS.pptx
RASHMI M G β€’7 views
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe... by Anmol Vishnu Gupta
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...
Anmol Vishnu Guptaβ€’28 views
Structure of purines and pyrimidines - Jahnvi arora (11228108), mmdu ,mullana... by jahnviarora989
Structure of purines and pyrimidines - Jahnvi arora (11228108), mmdu ,mullana...Structure of purines and pyrimidines - Jahnvi arora (11228108), mmdu ,mullana...
Structure of purines and pyrimidines - Jahnvi arora (11228108), mmdu ,mullana...
jahnviarora989β€’7 views
vitamine B1.pptx by ajithkilpart
vitamine B1.pptxvitamine B1.pptx
vitamine B1.pptx
ajithkilpartβ€’29 views
Discovery of therapeutic agents targeting PKLR for NAFLD using drug repositio... by Trustlife
Discovery of therapeutic agents targeting PKLR for NAFLD using drug repositio...Discovery of therapeutic agents targeting PKLR for NAFLD using drug repositio...
Discovery of therapeutic agents targeting PKLR for NAFLD using drug repositio...
Trustlifeβ€’146 views
A giant thin stellar stream in the Coma Galaxy Cluster by SΓ©rgio Sacani
A giant thin stellar stream in the Coma Galaxy ClusterA giant thin stellar stream in the Coma Galaxy Cluster
A giant thin stellar stream in the Coma Galaxy Cluster
SΓ©rgio Sacaniβ€’19 views
Factors affecting fluorescence and phosphorescence.pptx by SamarthGiri1
Factors affecting fluorescence and phosphorescence.pptxFactors affecting fluorescence and phosphorescence.pptx
Factors affecting fluorescence and phosphorescence.pptx
SamarthGiri1β€’7 views
A Ready-to-Analyze High-Plex Spatial Signature Development Workflow for Cance... by InsideScientific
A Ready-to-Analyze High-Plex Spatial Signature Development Workflow for Cance...A Ready-to-Analyze High-Plex Spatial Signature Development Workflow for Cance...
A Ready-to-Analyze High-Plex Spatial Signature Development Workflow for Cance...
InsideScientificβ€’115 views
ELECTRON TRANSPORT CHAIN by DEEKSHA RANI
ELECTRON TRANSPORT CHAINELECTRON TRANSPORT CHAIN
ELECTRON TRANSPORT CHAIN
DEEKSHA RANIβ€’11 views
Experimental animal Guinea pigs.pptx by Mansee Arya
Experimental animal Guinea pigs.pptxExperimental animal Guinea pigs.pptx
Experimental animal Guinea pigs.pptx
Mansee Aryaβ€’40 views
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F... by SwagatBehera9
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...
SwagatBehera9β€’5 views
λ³„ν—€λŠ” μ‚¬λžŒλ“€ 2023λ…„ 12μ›”ν˜Έ μ „λͺ…원 ꡐ수 자료 by sciencepeople
λ³„ν—€λŠ” μ‚¬λžŒλ“€ 2023λ…„ 12μ›”ν˜Έ μ „λͺ…원 ꡐ수 μžλ£Œλ³„ν—€λŠ” μ‚¬λžŒλ“€ 2023λ…„ 12μ›”ν˜Έ μ „λͺ…원 ꡐ수 자료
λ³„ν—€λŠ” μ‚¬λžŒλ“€ 2023λ…„ 12μ›”ν˜Έ μ „λͺ…원 ꡐ수 자료
sciencepeopleβ€’68 views
Assessment and Evaluation GROUP 3.pdf by kimberlyndelgado18
Assessment and Evaluation GROUP 3.pdfAssessment and Evaluation GROUP 3.pdf
Assessment and Evaluation GROUP 3.pdf
kimberlyndelgado18β€’10 views

Measuring the Precision of Multi-perspective Process Models

  • 1. Measuring the Precision of Multi-perspective Process Models Felix Mannhardt joint work with Massimiliano de Leoni, Hajo A. Reijers, Wil M.P. van der Aalst
  • 2. Precision Department of Mathematics and Computer Science PAGE 1 / 8 β€œFlower Model” lacking any precision B C A
  • 3. Precision of Multi-perspective Process Models Department of Mathematics and Computer Science PAGE 2 / 8 A π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐴 B π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐡 π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐡 > π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝐴 Existing work ignores added precision by multi-perspective rules / constraints
  • 4. Approach: Multi-perspective Precision Department of Mathematics and Computer Science PAGE 3 / 8 Multi-perspective Process Model (P) Fitting Event Log (L) Precision [0..1] INPUT OUTPUT APPROACH 𝒆 βˆˆπ‘³ 𝒐𝒃𝒔𝒆𝒓𝒗𝒆𝒅 𝑷 (𝒆) π’‘π’“π’†π’„π’Šπ’”π’Šπ’π’(𝑷, 𝑳) = 𝒆 βˆˆπ‘³ π’‘π’π’”π’”π’Šπ’ƒπ’π’† 𝑷(𝒆)
  • 5. Precision: Observed / Possible Behavior Department of Mathematics and Computer Science PAGE 4 / 8 C Id Event Loan obs pos 1 𝑒1 Handle Request 800 1 𝑒2 Simple Check - 1 𝑒3 Decide - 2 𝑒4 Handle Request 1800 2 𝑒5 Ext. Check - 2 𝑒6 Decide - 𝒐𝒃𝒔 𝑷 𝒆 𝟏 = { Handle Request } = 𝟏 𝒑𝒐𝒔 𝑷 𝒆 𝟏 = Handle Request = 𝟏 𝒔𝒕𝒂𝒕𝒆 𝒆 𝟏 = (< >, { }) C Id Event Loan obs pos 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 𝑒3 Decide - 2 𝑒4 Handle Request 1800 2 𝑒5 Ext. Check - 2 𝑒6 Decide - C Id Event Loan obs pos 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 1 1 𝑒3 Decide - 2 𝑒4 Handle Request 1800 2 𝑒5 Ext. Check - 2 𝑒6 Decide -𝒐𝒃𝒔 𝑷 𝒆 𝟐 = { Simple Check } = 𝟏 𝒑𝒐𝒔 𝑷 𝒆 𝟐 = Simple Check = 𝟏 𝒔𝒕𝒂𝒕𝒆 𝒆 𝟐 = (< 𝐻 >, { 𝐿 = 800 }) C Id Event Loan obs pos 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 1 1 𝑒3 Decide - 1 1 2 𝑒4 Handle Request 1800 2 𝑒5 Ext. Check - 2 𝑒6 Decide - 𝒐𝒃𝒔 𝑷 𝒆 πŸ‘ = { Decide } = 𝟏 𝒑𝒐𝒔 𝑷 𝒆 πŸ‘ = Decide = 𝟏 𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ‘ = (< 𝐻, 𝑆 >, { 𝐿 = 800 }) C Id Event Loan obs pos 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 1 1 𝑒3 Decide - 1 1 2 𝑒4 Handle Request 1800 1 1 2 𝑒5 Ext. Check - 2 𝑒6 Decide - 𝒐𝒃𝒔 𝑷 𝒆 πŸ’ = { Handle Request } = 𝟏 𝒑𝒐𝒔 𝑷 𝒆 πŸ’ = Handle Request = 𝟏 𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ’ = (<>, { }) C Id Event Loan obs pos 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 1 1 𝑒3 Decide - 1 1 2 𝑒4 Handle Request 1800 1 1 2 𝑒5 Ext. Check - 1 2 2 𝑒6 Decide -𝒐𝒃𝒔 𝑷 𝒆 πŸ“ = { Ext. Check } = 𝟏 𝒑𝒐𝒔 𝑷 𝒆 πŸ“ = Ext. Ceπ‘π‘˜, π‘†π‘–π‘šπ‘π‘™π‘’ πΆβ„Žπ‘’π‘π‘˜ = 𝟐 𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ“ = (< 𝐻 >, { 𝐿 = 1800 }) C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 1 1 𝑒3 Decide - 1 1 2 𝑒4 Handle Request 1800 1 1 2 𝑒5 Ext. Check - 1 2 2 𝑒6 Decide - 1 1 𝒐𝒃𝒔 𝑷 𝒆 πŸ” = Decide = 𝟏 𝒑𝒐𝒔 𝑷 𝒆 πŸ” = 𝐷𝑒𝑐𝑖𝑑𝑒 = 𝟏 𝒔𝒕𝒂𝒕𝒆 𝒆 πŸ” = (< 𝐻, 𝐸 >, { 𝐿 = 1800 }) C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 1 1 𝑒3 Decide - 1 1 2 𝑒4 Handle Request 1800 1 1 2 𝑒5 Ext. Check - 1 2 2 𝑒6 Decide - 1 1 6 7 𝒐𝒃𝒔 𝑷 𝒆 = { observed activities at state } 𝒑𝒐𝒔 𝑷 𝒆 = | π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ π‘Žπ‘π‘‘π‘–π‘£π‘–π‘‘π‘–π‘’π‘  π‘Žπ‘‘ π‘ π‘‘π‘Žπ‘‘π‘’ | 𝒔𝒕𝒂𝒕𝒆 𝒆 = π‘ π‘‘π‘Žπ‘‘π‘’ π‘œπ‘“ π‘‘β„Žπ‘’ π‘π‘Ÿπ‘œπ‘π‘’π‘ π‘  π‘šπ‘œπ‘‘π‘’π‘™ π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝑃, 𝐿 = π‘’βˆˆπΏ π‘œπ‘π‘ π‘’π‘Ÿπ‘£π‘’π‘‘ 𝑃(𝑒) = 6 π‘’βˆˆπΏ π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ 𝑃(𝑒) = 7 β‰ˆ 𝟎. πŸ–πŸ”
  • 6. Full Example for Model A & Model B Department of Mathematics and Computer Science PAGE 5 / 8 A C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 2 1 𝑒3 Decide - 1 1 2 𝑒4 Handle Request 1800 1 1 2 𝑒5 Extensive Check - 2 2 2 𝑒6 Decide - 1 1 3 𝑒7 Handle Request 1800 1 1 3 𝑒8 Simple Check - 2 2 3 𝑒9 Decide - 1 1 4 𝑒10 Handle Request 2500 1 1 4 𝑒11 Extensive Check - 1 2 4 𝑒12 Decide - 1 1 14 16 Model A π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝑃, 𝐿 = π‘’βˆˆπΏ π‘œπ‘π‘ π‘’π‘Ÿπ‘£π‘’π‘‘ 𝑃(𝑒) = 14 π‘’βˆˆπΏ π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ 𝑃(𝑒) = 16 β‰ˆ 𝟎. πŸ–πŸ•πŸ“ B C Id Event Loan 𝒐𝒃𝒔 𝑷 𝒑𝒐𝒔 𝑷 1 𝑒1 Handle Request 800 1 1 1 𝑒2 Simple Check - 1 1 1 𝑒3 Decide - 1 1 2 𝑒4 Handle Request 1800 1 1 2 𝑒5 Extensive Check - 2 2 2 𝑒6 Decide - 1 1 3 𝑒7 Handle Request 1800 1 1 3 𝑒8 Simple Check - 2 2 3 𝑒9 Decide - 1 1 4 𝑒10 Handle Request 2500 1 1 4 𝑒11 Extensive Check - 1 1 4 𝑒12 Decide - 1 1 14 14 Model B π‘π‘Ÿπ‘’π‘π‘–π‘ π‘–π‘œπ‘› 𝑃, 𝐿 = π‘’βˆˆπΏ π‘œπ‘π‘ π‘’π‘Ÿπ‘£π‘’π‘‘ 𝑃(𝑒) = 14 π‘’βˆˆπΏ π‘π‘œπ‘ π‘ π‘–π‘π‘™π‘’ 𝑃(𝑒) = 14 β‰ˆ 𝟏
  • 7. Evaluation on Road Fines Log 0.30 0.36 0.64 0.83 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Inductive Miner Inductive Miner & Rules Normative Model Normative Model & Rules ETC Precision Precision Fitness
  • 8. Summary Department of Mathematics and Computer Science PAGE 7 / 8 β€’ 1st precision measure for multi-perspective process models βˆ’ Fast to calculate βˆ’ Flexible framework βˆ’ Implemented in ProM β€’ Preliminary Evaluation βˆ’ Illustrative examples βˆ’ Real-life dataset with > 500,000 events Handle 750 Simple Decide Handle 1250 Ext. Decide Handle 5000 Simple Decide Handle 750 Simple Decide Handle 1500 Simple Decide precision [0..1]
  • 9. Department of Mathematics and Computer Science Image source: http://commons.wikimedia.org/wiki/File:Pictofigo_-_Idea.png Questions? Remarks? Ideas?