Labor Marketplace Applications: Human Computer Interaction KAIST KSE Uichin Lee 9/26/2011.

Slides:



Advertisements
Similar presentations
Interaction Design: Visio
Advertisements

A BPM Framework for KPI-Driven Performance Management
Using the Self Service BMC Helpdesk
02 | Define an Effective End-to-End Software Development Lifecycle Steven Borg | Co-founder & Strategist, Northwest Cadence Anthony Borton | ALM Consultant,
Software Development Languages and Environments. Programming languages High level languages are problem orientated contain many English words are easier.
Clipping Lists & Change Borders: Improving Multitasking Efficiency with Peripheral Information Design Mary Czerwinski George Robertson Desney Tan Microsoft.
Importing Transfer Equivalencies: How to Maximize Efficiency How Columbia College Office of Registrar improved productivity through third party solutions.
Introduction to Mechanized Labor Marketplaces: Mechanical Turk Uichin Lee KAIST KSE.
Chapter 14 Maintaining Information Systems Modern Systems Analysis and Design Seventh Edition Jeffrey A. Hoffer Joey F. George Joseph S. Valacich.
UI Standards & Tools Khushroo Shaikh.
James Tam User Centered Design Why User Centered Design is important Approaches to User Centered Design.
Crowdsourcing research data UMBC ebiquity,
CrowdFlow Integrating Machine Learning with Mechanical Turk for Speed-Cost-Quality Flexibility Alex Quinn, Ben Bederson, Tom Yeh, Jimmy Lin.
XP New Perspectives on Microsoft Office Word 2003 Tutorial 1 1 Microsoft Office Word 2003 Tutorial 1 – Creating a Document.
Principles and Methods
XP 1 Microsoft Office Word 2003 Tutorial 1 – Creating a Document.
The Visual Knowledge Builder: A Second Generation Spatial Hypertext Frank M. Shipman III Haowei Hsieh Preetam Maloor J. Michael Moore.
Bar|Scan ® Asset Inventory System The leader in asset and inventory management.
Task and Workflow Design I KSE 801 Uichin Lee. TurKit: Human Computation Algorithms on Mechanical Turk Greg Little, Lydia B. Chilton, Rob Miller, and.
Access 2007 ® Use Databases How can Microsoft Access 2007 help you manage a database?
Feeds Computer Applications to Medicine NSF REU at University of Virginia July 27, 2006 Paul Lee.
Computer Vision Systems for the Blind and Visually Disabled. STATS 19 SEM Talk 3. Alan Yuille. UCLA. Dept. Statistics and Psychology.
MIT HUMAN - COMPUTER INTERACTION Soylent A Word Processor with a Crowd Inside Michael Bernstein Greg Little, Rob Miller, David Karger, David Crowell, Katrina.
Mobile Multimodal Applications. Dr. Roman Englert, Gregor Glass March 23 rd, 2006.
TH-OCR NK. content introduction go to next page background assumptions overall structure chart IPO for overall structure dataflow diagram of overall structure.
XP M.S. Word Editing and Formatting a Document. XP New Perspectives on Microsoft Office 2007: Windows XP Edition2 Objectives Check spelling and grammar.
Web Science 2013 Paris, France May 1-5. Web Science emergent science of the people, organizations, applications, and of policies that shape and are shaped.
Systems Analysis – Analyzing Requirements.  Analyzing requirement stage identifies user information needs and new systems requirements  IS dev team.
UC Berkeley, Employment Services Candidate Gateway Tutorial External Applicants.
Microsoft Office Word 2003 Tutorial 1 Creating a Document.
Module 14: Configuring Print Resources and Printing Pools.
Enterprise & Intranet Search How Enterprise is different from Web search What to think about when evaluating Enterprise Search How Intranet use is different.
Workflow and SharePoint Presented by Ben Geers. Overview What is workflow? Windows Workflow Foundation How does workflow apply to SharePoint? WSS v3 vs.
Uichin Lee, Jihyoung Kim *, Eunhee Yi **, Juyup Sung, Mario Gerla * KAIST Knowledge Service Engineering * UCLA Computer Science ** LG UX R&D Lab
Microsoft Office 2007 Word Chapter 1 Creating and Editing a Word Document.
UC Berkeley Employment Services Candidate Gateway Tutorial Internal Applicants.
Design IS 101Y/CMSC 101 Computational Thinking and Design Tuesday, October 15, 2013 Carolyn Seaman University of Maryland, Baltimore County.
A haptic presentation of 3D objects in virtual reality for the visually disabled MSc Marcin Morański Professor Andrzej Materka Institute of Electronics,
Exploration Seminar 3 Human Computation Roy McElmurry.
Labor Marketplace Applications KAIST KSE Uichin Lee.
Crowdsourcing: Ethics, Collaboration, Creativity KSE 801 Uichin Lee.
Software Project Planning Defining the Project Writing the Software Specification Planning the Development Stages Testing the Software.
TOPIC CENTRIC QUERY ROUTING Research Methods (CS689) 11/21/00 By Anupam Khanal.
Task and Workflow Design in Human Computation KSE 652 Social Computing System Design and Analysis Uichin Lee.
Computing Old and New: Labor Markets for Data David Alan Grier Center for International Science & Technology Policy George Washington University.
CrowdSearch: Exploiting Crowds for Accurate Real-Time Image Search on Mobile Phones Original work by Tingxin Yan, Vikas Kumar, Deepak Ganesan Presented.
1 CMPT 275 High Level Design Phase Modularization.
MIT 6.893; SMA 5508 Spring 2004 Larry Rudolph Lecture Introduction Sketching Interface.
Creating a Corpus for A Conversational Assistant for Everyday Tasks Henry Kautz, Young Song, Ian Pereira, Mary Swift, Walter Lasecki, Jeff Bigham, James.
Input Design Lecture 11 1 BTEC HNC Systems Support Castle College 2007/8.
Connecting with Computer Science2 Objectives Learn how software engineering is used to create applications Learn some of the different software engineering.
EUD at Scale Andrew Begel, Microsoft Research. Classic Human Programming Madlibs
Copyright © 2006 – Brad A. Myers Answering Why and Why Not Questions in User Interfaces Brad Myers, David A. Weitzman, Andrew J. Ko, and Duen Horng (“Polo”)
Importance of user interface design – Useful, useable, used Three golden rules – Place the user in control – Reduce the user’s memory load – Make the.
Learning Objectives Understand the concepts of Information systems.
UFCFY5-30-1Multimedia Studio Scripting for Interactive Media Times Table Quiz This will contribute towards your online portfolio for this module.
Matt Tempia Chris Thomas Thomas Taylor Jeffrey Tansey.
Systems and User Interface Software. Types of Operating System  Single User  Multi User  Multi-tasking  Batch Processing  Interactive  Real Time.
CIS-NG CASREP Information System Next Generation Shawn Baugh Amy Ramirez Amy Lee Alex Sanin Sam Avanessians.
Some of the utilities associated with the development of programs. These program development tools allow users to write and construct programs that the.
Computer Fundamentals
Input and output devices for visually impaired users
SIE 515 Design Evaluation Lecture 7.
Testing Multimedia Products
A crowd-sourcing framework for automated visualization evaluation
CAPE Internal Assessment
Presented by Nguyen Kieu Oanh (M.A)
Successful Website Accessibility Testing
MyWritingLabPlus for Economics
Presentation transcript:

Labor Marketplace Applications: Human Computer Interaction KAIST KSE Uichin Lee 9/26/2011

Soylent: A Word Processor with a Crowd Inside Michael Bernstein, Greg Little, Rob Miller, David Karger, David Crowell, Katrina Panovich MIT CSAIL Björn Hartmann, Mark Ackerman UC Berkeley University of Michigan UIST 2010 Part of slides are from

Soylent: a word processing interface Uses crowd contributions to aid complex writing tasks.

Challenges for Programming Crowds The authors have interacted with ~9000 Turkers on ~2000 different tasks Key Problem: crowd workers often produce poor output on open-ended tasks – High variance of effort: Lazy Turker vs. Eager Beaver – Errors made by Turkers; e.g., “Of Mice and Men”  “Of Mice” or calling it a movie.. 30% Rule: ~30% of the results from open-ended tasks will be unsatisfactory Solution: find-fix-verify

(at least 20% of the Turkers agree upon) (e.g., given to 10 Tuckers) (e.g., given to 5 Tuckers)

Find-Fix-Verify Discussion Why split Find and Fix? – Force Lazy Turkers (tend to choose easiest work) to work on a problem of our choice – Allows us to merge work completed in parallel Why add Verify? – Quality rises when we place Turkers in productive tension – Allows us to trade off lag time with quality Timeout at each stage for responsive operations (e.g., minutes)

Evaluation Implementation – Microsoft World plug-in using MS Visual Studio Tools for Office (VSTO) and Windows Presentation Foundation (WPF) – TurKit Mechanical Turk toolkit Evaluation: – Shortn, Crowdproof, Human Macro Metrics: quality, delay, cost

Shortn Evaluation Setting: – Find: 6-10 workers ($0.08 per Find) – Fix/Verify: 3-5 workers ($0.05 per Fix, $0.04 per Verify) – Delay: wait time, work time Results: – 78%-90% reduction.. – Wait time of all stages: median 18.5 minutes – Work time: median 118 seconds – Average costs: $1.41 ($ $ $0.38) – Caveats: some modifications are grammatically appropriate, but stylistically incorrect

Crowdproof Evaluation Tested 5 input texts – Manually labeled all spelling, grammatical and style errors in each of the five inputs (total 49 errors) Ran Crowdproof w/ a 20-minute stage timeout; and measure how many corrections Crowdproof makes – Task setup: Find: $0.06, Fix: $0.08, Verify: $0.04 Soylent: 67% vs. MS Word (grammar check): 30% – Combined: 82%

Human Macro Evaluation Give 5 prompts (Input and Output) to users (CS students, Admin, Author) Ask to generate descriptions (Request) used for Human Macro tasks

Discussion Delay in interface outsourcing: minutes to hours.. Privacy? Confidential documents? Legal ownership? Lack of domain knowledge or shared context to usefully contribute

VizWiz: Nearly Real-time Answers to Visual Questions Jeffrey P. Bigham, Chandrika Jayant, Hanjie Ji, Greg Little, Andrew Miller, Robert C. Miller, Robin Miller, Aubrey Tatarowicz, Brandyn White, Samual White, and Tom Yeh. UIST Part of slides are from:

VizWiz “automatic and human-powered services to answer general visual questions for people with visual impairments.” Lets blind people use mobile phone to: 1. Take a photo 2. Speak a question 3. Receive multiple spoken answers

Motivation Current technology uses automatic approaches to help blind people access visual information – Optical Character Recognition (OCR) – Ex) Kurzweil knfbReader: ~$1,000 Problems: error-prone, limited in scope, expensive – Ex: OCR cannot read graphic labels, handwritten menu, street sign

Motivation Solution: ask real people Can phrase questions naturally – “What is the price of the cheapest salad?” vs. OCR reading the entire menu Feedback – Real people can guide blind people to take better photos Focus on blind people’s needs, not current technology

Human-Powered Services Problem: Response latency Solution: quikTurkit (and some tricks) – “First attempt to get work done by web based workers in nearly real-time” – Maintain a pool of workers to answer questions

quikTurkit “requesters create their own web site on which Mechanical Turk workers answer questions.” “answers are posted directly to the requester’s web site, which allows [them]... to be returned before an entire HIT is complete.” “workers are required to answer multiple previously-asked questions to keep workers around long enough to possibly answer new questions”

Answering Site

Deployment Setting: 11 blind iPhone users quikTurkit: – Reward range: $0.01 (1/2 of jobs), $0.02/3 (1/2) Results: 86.6% of first answers “correct” – Average of 133.3s latency for first answer Problems: Photos too dark or too blurry and thus unanswerable. VizWiz 2.0 detects and alerts users if photo is too dark or blurry

VizWiz: LocateIt Combine VizWiz with computer vision to help blind people locate objects LocateIt mobile: Sensor (zoom and filter) and sonification modules Web interface (angle) (distance)

Future Work Expand worker pools beyond Mechanical Turk (e.g. social network) Reduce cost by using game, volunteers, friends Improve interface to make photo-taking easier for blind people Combine automatic approaches to improve delay

Discussion Resource usage vs. delay (wasting resources for better responses or near real-time services?) – Any better approach than quikTurkit? Quality control? How do we make sure that the workers correctly identified the photos? – How do systems accept/reject the submissions? (if it’s becoming a large scale service?) Other application scenarios with quikTurkit? Adding inertial sensors to LocateIt?