Crowd-based mining of reusable process model patterns Carlos Rodríguez, Florian Daniel, Fabio Casati BPM 2014, September 9th 2014, Eindhoven, The Netherlands.

Slides:



Advertisements
Similar presentations
A Workflow Engine with Multi-Level Parallelism Supports Qifeng Huang and Yan Huang School of Computer Science Cardiff University
Advertisements

Xchange-Park: A Crowd-sourcing based parking reservation system Aakash Therani Ankit Jasuja Manish Shah
Foundational Objects. Areas of coverage Technical objects Foundational objects Lessons learned from review of Use Case content Simple Study Simple Questionnaire.
Incentivize Crowd Labeling under Budget Constraint
CrowdER - Crowdsourcing Entity Resolution
‘s Overload Tolerant Design Exacerbates Failure Detection and Recovery Florin Dinu T. S. Eugene Ng Rice University.
Crowdsourcing ontology engineering Elena Simperl Web and Internet Science, University of Southampton 11 April 2013.
Applying Crowd Sourcing and Workflow in Social Conflict Detection By: Reshmi De, Bhargabi Chakrabarti 28/03/13.
Utilizing a Compositional System Knowledge Framework for Ontology Evaluation: A Case Study on BioSTORM H.Hlomani, M.G.Gillespie, D.Kotowski, D. A. Stacey.
Collective Intelligence Dr. Frank McCown Intro to Web Science Harding University This work is licensed under a Creative Commons Attribution-NonCommercial-
MI021/CS021: Computers in Management April 24, 2009 Peer Production, Social Media, and Web 2.0 Prof. John Gallaugher written case &
Research topics Semantic Web - Spring 2007 Computer Engineering Department Sharif University of Technology.
ODE: Ontology-assisted Data Extraction WEIFENG SU et al. Presented by: Meher Talat Shaikh.
Crowdsourcing research data UMBC ebiquity,
Job Analysis & Its Components Job analysis: Job analysis: The process of describing and recording many aspects or elements of the job. The outcome of job.
Aligning Business Processes to SOA B. Ramamurthy 6/16/2015Page 1.
Support for Automatic Workflow Composition in Semantic Grid Environemnt Tomasz Gubała, Marian Bubak, Maciej Malawski Institute of Computer Science and.
Discovering Coordination Patterns using Process Mining Prof.dr.ir. Wil van der Aalst Eindhoven University of Technology Department of Information and Technology.
1/31 CS 426 Senior Projects Chapter 1: What is UML? Chapter 2: What is UP? [Arlow and Neustadt, 2005] January 22, 2009.
Web Archive Information Retrieval Miguel Costa, Daniel Gomes (speaker) Portuguese Web Archive.
Chapter 10: Architectural Design
Presenters: Title:. CONTENTS What is Crowdsourcing? How Crowdsourcing works? Types of Crowdsourcing Applications of Crowdsourcing Benefits & Problems.
Students: Ilya Paskhover, Itay Gal Supervisors: Oleg Rokhlenko, Nadav Golbandi.
Human Computation and Crowdsourcing Uichin Lee May 8, 2011.
Špindlerův Mlýn, Czech Republic, SOFSEM Semantically-aided Data-aware Service Workflow Composition Ondrej Habala, Marek Paralič,
Crowdsourcing. What is it? Image credit: Alexander Kesselaar.
1 Yolanda Gil Information Sciences InstituteJanuary 10, 2010 Requirements for caBIG Infrastructure to Support Semantic Workflows Yolanda.
CHAPTER 4 Marketing Information and Research: Analyzing the Business Environment Off-line and Online M A R K E T I N G.
1 Intelligent Systems ISCRAM 2013 Validating Procedural Knowledge in the Open Virtual Collaboration Environment Gerhard Wickler AIAI, University.
DATA-CENTERED CROWDSOURCING WORKSHOP PROF. TOVA MILO SLAVA NOVGORODOV TEL AVIV UNIVERSITY 2014/2015.
Name : Emad Zargoun Id number : EASTERN MEDITERRANEAN UNIVERSITY DEPARTMENT OF Computing and technology “ITEC547- text mining“ Prof.Dr. Nazife Dimiriler.
GLOSSARY COMPILATION Alex Kotov (akotov2) Hanna Zhong (hzhong) Hoa Nguyen (hnguyen4) Zhenyu Yang (zyang2)
Crowdsourcing in Practice The act of taking a job traditionally performed by a designated agent (usually an employee) and outsourcing it to an undefined,
Crowdsourcing with Multi- Dimensional Trust Xiangyang Liu 1, He He 2, and John S. Baras 1 1 Institute for Systems Research and Department of Electrical.
Overview Crowd-Powered Systems Alexander J. Quinn January 15, 2015.
WebMining Web Mining By- Pawan Singh Piyush Arora Pooja Mansharamani Pramod Singh Praveen Kumar 1.
1 Project  By group (1-4 students, formed on your own)  Each group  select 1 topic  hand in 1 report (page limit: 5 pages; due: July 10)  give 1 presentation.
Min Chen, and Yuhong Yan Concordia University, Montreal, Canada Presentation at ICWS 2012 June 24-29, 2012, Hawaii (Honolulu), USA Redundant Service Removal.
Markup and Validation Agents in Vijjana – A Pragmatic model for Self- Organizing, Collaborative, Domain- Centric Knowledge Networks S. Devalapalli, R.
Transfer Learning Motivation and Types Functional Transfer Learning Representational Transfer Learning References.
Procedures for managing workflow components Workflow components: A workflow can usually be described using formal or informal flow diagramming techniques,
Document Clustering for Forensic Analysis: An Approach for Improving Computer Inspection.
Crowdsourcing Nathan McFarland Isaac Nichols February 26 th, 2013.
11 CORE Architecture Mauro Bruno, Monica Scannapieco, Carlo Vaccari, Giulia Vaste Antonino Virgillito, Diego Zardetto (Istat)
1 Limitations of BLAST Can only search for a single query (e.g. find all genes similar to TTGGACAGGATCGA) What about more complex queries? “Find all genes.
Chapter 9: Tapping the Crowd for Fast Innovation ISTO SIPILÄ.
The Rise of Crowdsourcing in Management Research Organized by: Yuqing (Ching) Ren Natalia Levina August 9, 2010 Academy of Management Annual Meeting.
Instance Discovery and Schema Matching With Applications to Biological Deep Web Data Integration Tantan Liu, Fan Wang, Gagan Agrawal {liut, wangfa,
Bringing Order to the Web : Automatically Categorizing Search Results Advisor : Dr. Hsu Graduate : Keng-Wei Chang Author : Hao Chen Susan Dumais.
4- Copyright © 2015 Pearson Education, Inc Job Analysis and the Talent Management Process.
Semi-Supervised Recognition of Sarcastic Sentences in Twitter and Amazon -Smit Shilu.
WI2003 Automatic Composition of Web Service Workflows Using a Semantic Agent Jarmo Korhonen Helsinki University of Technology 15 October 2003.
Implementation of Classifier Tool in Twister Magesh khanna Vadivelu Shivaraman Janakiraman.
Bossa: A platform for distributed thinking David P. Anderson UC Berkeley Space Sciences Lab 11 Oct 2007.
Business process management (BPM)
Job Analysis Chapter 4 Md. Al-Amin.
Data-Centered Crowdsourcing Workshop
C2A: Crowd Consensus Analytics for Virtual Colonoscopy
Unit – 5 JAVA Web Services
Business process management (BPM)
The 2007 Winter Conference on Business Intelligence
Many hands make light work…
Elsevier Activity Range
Tim Sheerman-Chase, Eng-Jon Ong and Richard Bowden
Lesson 1: Introduction to Trifacta Wrangler
Lesson 1: Introduction to Trifacta Wrangler
A General Approach to Real-time Workflow Monitoring
HW #1 Think about your current job or a past job and answer
Job Analysis Chapter 4 Md. Al-Amin.
Presentation transcript:

Crowd-based mining of reusable process model patterns Carlos Rodríguez, Florian Daniel, Fabio Casati BPM 2014, September 9th 2014, Eindhoven, The Netherlands

Baya (Rodríguez et. al., 2014): An extension for Yahoo! Pipes that interactively recom- mends mashup model patterns during pipe modeling Context and motivation (1) 2

Context and motivation (2) Example patternKnowledge base Issues: Identification of the right support threshold values Large number of patterns produced Noise (useless patterns) Giving meaning to patterns Difficulty in finding patterns from small datasets 3

Crowdsourcing Amazon Mechanical Turk Innocentive “Crowdsourcing is the act of taking a job traditionally performed by a designated agent (usually an employee) and outsourcing it to an undefined, generally large group of people in the form of an open call.” (Howe, 2006) 4 CrowdFlower

Crowdsourcing models and workflow Crowdsourcing models: Workflow for crowdsourcing a task: Market place Contest Auction 5 Design and publish task Search and inspect tasks Search and inspect tasks Pre-select workers Execute task Validate results Start End Crowdsourcer Worker

Requirements Description: xxxxxx Tags: xxxxx, xxx, xxx X R1: Qualification tests R2: Mashup model representation R3: Pattern description R4: Input checking R5: Use of redundancy 6

The approach Fig. 1: Our approach to crowd-based pattern mining with CrowdFlower 7

Fig. 2: Pre-selection questionnaire used to assess worker’s acquaintance with Yahoo! Pipes Pre-selection of workers 8

Task design Fig. 3: Task design for the selection, description and rating of mashup model patterns 9

H3 - Applicability Crowd-based pattern mining outperforms machine- based pattern mining for small datasets H1 - Effectiveness It is possible to mine reusable mashup model patterns from mashup models by crowdsourcing the identification of patterns H2 - Value Model patterns identified by the crowd contain more domain knowledge than automatically mined patterns Experiment design (1) vs. Crowd Machine 10

Experiment design (2) Dataset for machine 997 pipes with 11.1 components and 11.0 connectors on average (in JSON format) Dataset for crowd 40 pipes randomly selected from the 997 pipes above (including both the image of the pipes and their JSON representation) Algorithms used Machine 997, Machine 40 and Crowd 40 Crowd 40 settings USD 0.10 per task, 3 judgments per pipe, 300 seconds per task Machine settings We run the machine algorithms using different minimum support values 11

Results (1) Fig. 4: Task instances and patterns in Crowd 40 Fig. 5: Number of patterns produced Machine 997 Crowd 40 Machine 40 Crowd 40 H1 - Effectiveness 12

Results (2) Fig. 6: Average pattern size Fig. 7: Size distribution of patterns Machine 997 Crowd 40 Machine 40 Crowd 40 Machine 997 Machine 40 Crowd 40 H2 - Value H3 - Applicability 13

Discussion and analogy with BPM Rewards: USD 0.10 vs. USD 0.05 Strong sensitivity to task design Conclusion: KISS (Keep it Simple, Stupid) Focus on the validation of collected data At an abstract level, BP models are not very dissimilar to pipes Control flow based vs. Data flow based Structure of the model patterns Lessons learned: Business process models vs. Mashups models: 14

Future work Work with larger datasets Experiment with more, different and wider ranges of rewards to understand the effects on the quality of the resulting patterns Crowdsource the mining of BP model patterns 15

Thanks

References 1.Carlos Rodríguez, Soudip Roy Chowdhury, Florian Daniel, Hamid Motahari Nezhad, Fabio Casati. Assisted mashup development: On the discovery and recommendation of mashup composition knowledge. Book chapter. Web services foundation, Springer (2014) 2.Jeff Howe. Crowdsourcing: Why the power of the crowd is driving the future of business. URL: html (2006)