L C SL C S Supporting Technology for Group Interaction Howard Shrobe MIT AI Lab Oxygen Workshop, January, 2002.

Slides:



Advertisements
Similar presentations
CONCEPTUAL WEB-BASED FRAMEWORK IN AN INTERACTIVE VIRTUAL ENVIRONMENT FOR DISTANCE LEARNING Amal Oraifige, Graham Oakes, Anthony Felton, David Heesom, Kevin.
Advertisements

Facilitated by Joanne Fraser RiverSystems
Lecture 8: Three-Level Architectures CS 344R: Robotics Benjamin Kuipers.
Describing Process Specifications and Structured Decisions Systems Analysis and Design, 7e Kendall & Kendall 9 © 2008 Pearson Prentice Hall.
1.Data categorization 2.Information 3.Knowledge 4.Wisdom 5.Social understanding Which of the following requires a firm to expend resources to organize.
I-Room : Integrating Intelligent Agents and Virtual Worlds.
Collaborative Work Systems, Inc CWS Collaborative Work Systems, Inc Geo-Docent: Improving Human, Team, and Organizational Performance with Geographically.
Component-Based Software Engineering Oxygen Paul Krause.
1 Intelligent Agents Software analog to human agents real estate agent, librarian, salesperson Perform tasks individually, or in collaboration Static and.
Chapter 19: Network Management Business Data Communications, 4e.
Object-Oriented Analysis and Design
Think. Learn. Succeed. Aura: An Architectural Framework for User Mobility in Ubiquitous Computing Environments Presented by: Ashirvad Naik April 20, 2010.
Ambient Computational Environments Sprint Research Symposium March 8-9, 2000 Professor Gary J. Minden The University of Kansas Electrical Engineering and.
Knowledge Acquisitioning. Definition The transfer and transformation of potential problem solving expertise from some knowledge source to a program.
An Active Events Model for Systems Monitoring Philip Gross Columbia University Programming Systems Lab Director: Gail Kaiser.
AceMedia Personal content management in a mobile environment Jonathan Teh Motorola Labs.
Smart Space & Oxygen CIS 640 Project By Usa Sammpun
© 2004, The Trustees of Indiana University 1 OneStart Workflow Basics Brian McGough, Manager, Systems Integration, UITS Ryan Kirkendall, Lead Developer.
1 I/O Management in Representative Operating Systems.
Behavior-Based Artificial Intelligence Pattie Maes MIT Media-Laboratory Presentation by: Derak Berreyesa UNR, CS Department.
Semantic Web and Web Mining: Networking with Industry and Academia İsmail Hakkı Toroslu IST EVENT 2006.
Smart Learning Services Based on Smart Cloud Computing
Semantic Web Technologies Lecture # 2 Faculty of Computer Science, IBA.
LÊ QU Ố C HUY ID: QLU OUTLINE  What is data mining ?  Major issues in data mining 2.
CASE Tools And Their Effect On Software Quality Peter Geddis – pxg07u.
Presented to: By: Date: Federal Aviation Administration Enterprise Information Management SOA Brown Bag #2 Sam Ceccola – SOA Architect November 17, 2010.
Learner Modelling in a Multi-Agent System through Web Services Katerina Kabassi, Maria Virvou Department of Informatics, University of Piraeus.
Katanosh Morovat.   This concept is a formal approach for identifying the rules that encapsulate the structure, constraint, and control of the operation.
L C SL C S Supporting Technology for Group Interaction Howard Shrobe MIT AI Lab Oxygen Workshop, January, 2002.
1 Shawlands Academy Higher Computing Software Development Unit.
SensIT PI Meeting, January 15-17, Self-Organizing Sensor Networks: Efficient Distributed Mechanisms Alvin S. Lim Computer Science and Software Engineering.
An Overview of MPEG-21 Cory McKay. Introduction Built on top of MPEG-4 and MPEG-7 standards Much more than just an audiovisual standard Meant to be a.
© 2007 Tom Beckman Features:  Are autonomous software entities that act as a user’s assistant to perform discrete tasks, simplifying or completely automating.
Sharad Oberoi and Susan Finger Carnegie Mellon University DesignWebs: Towards the Creation of an Interactive Navigational Tool to assist and support Engineering.
1 The Software Development Process  Systems analysis  Systems design  Implementation  Testing  Documentation  Evaluation  Maintenance.
1 COMPSCI 110 Operating Systems Who - Introductions How - Policies and Administrative Details Why - Objectives and Expectations What - Our Topic: Operating.
Describing Process Specifications and Structured Decisions Systems Analysis and Design, 7e Kendall & Kendall 9 © 2008 Pearson Prentice Hall.
Patterns and Reuse. Patterns Reuse of Analysis and Design.
Chapter © 2012 Pearson Education, Inc. Publishing as Prentice Hall.
ARTIFICIAL INTELLIGENCE [INTELLIGENT AGENTS PARADIGM] Professor Janis Grundspenkis Riga Technical University Faculty of Computer Science and Information.
Subtask 1.8 WWW Networked Knowledge Bases August 19, 2003 AcademicsAir force Arvind BansalScott Pollock Cheng Chang Lu (away)Hyatt Rick ParentMark (SAIC)
Advanced Computer Networks Topic 2: Characterization of Distributed Systems.
FOREWORD By: Howard Shrobe MIT CS & AI Laboratory
Dynamic Domain Architectures for Model-based Autonomy Bob Laddaga Howard Shrobe Brian C. Williams (PI) MIT Artificial Intelligence Lab Space Systems Lab.
A Context Model based on Ontological Languages: a Proposal for Information Visualization School of Informatics Castilla-La Mancha University Ramón Hervás.
1 What is OO Design? OO Design is a process of invention, where developers create the abstractions necessary to meet the system’s requirements OO Design.
The Second Life of a Sensor: Integrating Real-World Experience in Virtual Worlds using Mobile Phones Mirco Musolesi, Emiliano Miluzzo, Nicholas D. Lane,
L C SL C S Reactive and Responsive Intelligent Environments Kevin Quigley aire group MIT AI Lab.
The Software Development Process
L C SL C S The Intelligent Room’s MeetingManager: A Look Forward Alice Oh Stephen Peters Oxygen Workshop, January, 2002.
Chapter 4 Decision Support System & Artificial Intelligence.
CSC480 Software Engineering Lecture 10 September 25, 2002.
IT and Network Organization Ecommerce. IT and Network Organization OPTIMIZING INTERNAL COLLABORATIONS IN NETWORK ORGANIZATIONS.
Integrated Knowledge System on Climate Change Adaptation Conceptual & Technological Framework OneWorld South Asia December 2008.
Project Management Training
L C SL C S Metaglue: Overview Of Current Challenges Krzysztof Gajos Oxygen Workshop, January, 2002.
1 The Software Development Process ► Systems analysis ► Systems design ► Implementation ► Testing ► Documentation ► Evaluation ► Maintenance.
Erik Jonsson School of Engineering and Computer Science The University of Texas at Dallas Cyber Security Research on Engineering Solutions Dr. Bhavani.
Semantic Web in Context Broker Architecture Presented by Harry Chen, Tim Finin, Anupan Joshi At PerCom ‘04 Summarized by Sungchan Park
Slide no 1 Cognitive Systems in FP6 scope and focus Colette Maloney DG Information Society.
CS10K Community Facilitators and Social Learning Team Meeting January 14, 2013 Portland, OR.
Human Computer Interaction Lecture 21 User Support
A Context Framework for Ambient Intelligence
Chapter 19: Network Management
Course Outcomes of Object Oriented Modeling Design (17630,C604)
Unified Modeling Language
Enabling Collaboration with IT
Developing Innovative Unified Communications Applications
MANAGING KNOWLEDGE FOR THE DIGITAL FIRM
Presentation transcript:

L C SL C S Supporting Technology for Group Interaction Howard Shrobe MIT AI Lab Oxygen Workshop, January, 2002

L C SL C S Three Motivating Questions Why do organizations repeat their past mistakes? Why do organizations fail to notice impending opportunities and crises? Why do projects never complete on time? Because they fail to capture, organize and disseminate the information which groups exchange in their daily work lives. Because capturing such information is unnatural –Interrupts your actual task –Has value only in the future –Has value to somebody else. Because capturing it costs more than it’s worth! This is ironic because designers want to discuss their creations.

L C SL C S A Three Pronged Attack on Collaborative Knowledge Management Ubiquitous human-centered, perceptually enabled environments. An adaptive infrastructure that understands the context and content of problem solving discourse (at least a little) –It accurately indexes all information. –It distributes information to those who can use it. –It can find resources (information and people) to help a collaboration. Tools informed by an understanding of the domain and the organization. –An “almost expert” (apprentice) system *Which knows when to ask questions: *What it doesn’t understand is often what’s most important *What isn’t obvious is what needs to be documented and distributed Documentation almost for free by embedding a helpful computational system in our normal work flow

L C SL C S Team Based Collaboration Is oriented around shared goals Is opportunistic and information driven Requires information to be distributed to those who need it –And not to those who don’t Involves forming, deliberating about and executing shared plans –Each step of the plan sets the context for how people (should) interact –Plans are specific to the task and the organization

L C SL C S An Example of Group Interaction and Shared Plans Clarify Interface Propose Code Document Brainstor m Critique The side-tracker interface would be much clearer if it worked this way! I think I can make it work that way, but does it affect anything else! That should work, as long as it doesn’t break the special hack for managers. I’ll look into that, you all can start the coding and doc updates Plan

L C SL C S A Meeting Web Video/Audio Transcript (quicktime movie) Position 1 Start time; End time: Position 2 Start time; End time: Discourse Structure Issue 1 Start time; End time: Issue 2 Start time; End time: Supporting Argument Start time; End time: Refuting Argument Start time; End time: Agenda Item Start time; End time: Agenda Item Start time; End time: Agenda Item Start time; End time: Meeting Structure Commitment Start time; End time: Who: Deadline Discourse Commitment Start time; End time: Who: Deadline Discourse Action Items Commitment Start time; End time: Who: Deadline Discourse People

L C SL C S Information nodes: Goals of the project Proposed methods for the goals Arguments in favor and against methods Documents supporting these arguments Node format: Much of the content is opaque to the system (e.g. Multi-media fragments) Some slots are understood by the system And, short natural language annotations can be attached, parsed and understood by START Formal representations are also possible Relational information: Links showing the relationships between the information nodes Link types are meaningful to the system Organizational descriptions: Resource descriptions including capabilities, roles and interests (for personnel) Process goals and plans Meeting Manager Plan KB Goal Plan Knowledge Base Haystacks Knowledge acquisition Facilitator Shared Plans Shared Info Shared Semantic Web Inference Resource Discovery web The Oxygen Collaboration Infrastructure: Knowledge Based Collaboration Webs

L C SL C S Plan KB Goal Plan Knowledge Base Haystacks Knowledge acquisition Facilitator Shared Plans Shared Info Shared Semantic Web Inference Resource Discovery Facilitator Agent A Knowledge-Based Web Helps People Collaborate By acting as an assistant to meeting facilitators it can help to capture important information such as: –Issue, positions, arguments for and against positions –Commitments, action items –Video and audio transcripts of meetings By understanding the role of individuals within the organization it can help to decide who should see what information. By understanding the technical capabilities of individuals it can help to match “who can do what” to “what is required”.

L C SL C S A Collaboration Web Helps by Making Simple Inferences Oxygen deduces new information using: –The background knowledge base –The types of the links –The slots of node structures that are understood –The semantic content of the node annotations –Descriptions of the resources and people in the organization Oxygen routes information and discussion topics to those whose: –Organizational role requires it –Interests suggest it –Capabilities and skills might be useful Oxygen posts new goals and initiates new discussion processes to address these goals. Oxygen supports human interactions in a manner relevant to the organization’s problem solving context

L C SL C S Oxygen Makes This Much Easier by Bringing Computers Into Our World Microphone Array Tracking Cameras Video Displays Pointing Camera Pointing Camera

Summarizing The Challenges: A Unified Infrastructure For Group Interactions

L C SL C S Metaglue: A Platform for Multi-modal HCI Agents Are the Basic Building Block of Meta-glue An Agent Is a Software Object That Acts on Behalf of Some Entity in the Real World –Agents Are Distributed Objects That Run in Multiple Computers and Virtual Machines and That Communicate by Direct Message Passing –Agents Find One Another Through a Catalog That Provides Resource Discovery Capabilities –Many Basic Agents Control and Represent the State of Devices Agents may communicate directly Agents Are Clustered Into Societies –A Society Is a Collection of Agents –All Agents in a Society Act on Behalf of a Common Real-world Entity Such As a Space or a Person. Events Are Qualitative Changes in Significant Properties of the World Events Are Shared Through a Publish & Subscribe Broadcast Medium

L C SL C S Structuring of Agent Societies Projector agent Resource manager VCR agent Secretary Browser agentSociety agent agent Secretary Resource manager Society agent Conference room societyMy society Catalog

L C SL C S 5 Keys Challenges for Adaptive Interfaces Providing a practical level of knowledge representation that enables groups interactions and grounding in the real world of space and time Providing services in a multi-user environment while making optimal use of the currently available resources Recovering from equipment failures, information attacks, etc. Coordinating and fusing information from many sensors and modalities Capitalizing and recognizing context Maintaining Security and Privacy and trading these off against other goals

L C SL C S Challenge 1: Grounding in Real-World Semantics We want to build applications that service many individuals and groups of individuals These people will move among many physical spaces The devices and resources they use change as time progresses The context shifts during interactions The relevant information base evolves as well.

L C SL C S Challenge 1: Grounding in Real-World Semantics We want to build applications that service many individuals and groups of individuals These people will move among many physical spaces The devices and resources they use change as time progresses The context shifts during interactions The relevant information base evolves as well. The system is required to respond dynamically

L C SL C S Research Agenda: Knowledge Representations People –Interests, skills, responsibilities, organizational role Organizations –Members, structure Spaces –Location –Subspaces –Devices and resources Resources Information nodes –Topic area, place in ontology, format Services –Methods, parameter bindings, resource requirements Agents –Capabilities, society, acting on behalf of whom Events –E.g. Person identification, motion into a new region of space, gestures –Qualitative Changes in any of the properties in the KR

L C SL C S Challenge 2: Adaptive Resource Management In most systems, applications are written in terms of specific resources –(e.g. The left projector in Michael’s office, or worse yet, a physical address). This is in conflict with –Portability across physical contexts –Changes in equipment availability across time –Multiple applications demanding similar resources –Need to take advantage of new resources –Need to integrate mobile devices as they migrate into a space –Need to link two or more spaces What is required is a more abstract approach to resources in which no application needs to be tied to a specific device.

L C SL C S Abstract Service Control Parameters User’s Utility Function The binding of parameters has a value to the user Resource 1,1 Resource 1,2 Resource 1,j Each method requires different resources The system selects the method which maximizes net benefit User requests A service with certain parameters Resource Cost Function The resources Used by the method Have a cost Net benefit Each method binds the settings of The control parameters in a different way Method 1 Method 2 Method n Each service can be Provided by several Methods The system adapts by having many plans for each service

L C SL C S Security and Privacy Issues Are Addressed by Factoring in the Cost of Violating a Security or Privacy Policy Resource Cost Function Resource 1,1 Resource 1,2 Resource 1,j Abstract Service User’s Utility Function Net Benefit Method 1 Method 2 Method n Security Policies Challenge 6: Security and Privacy Is Addressed by This Infrastructure

L C SL C S Challenge 3: Robustness and Recovery From Failures The intelligent room renders services by translating them into plans involving physical resources –Physical resources have known failure modes Each plan step accomplishes sub-goals needed by succeeding steps –Each sub-goal has some way of monitoring whether it has been accomplished –These monitoring steps are also inserted into the plan If a sub-goal fails to be accomplished, model-based diagnosis isolates and characterizes the failure A recovery is chosen based on the diagnosis –It might be as simple as “try it again”, we had a network glitch –It might be “try it again, but with a different selection of resources” –It might be as complex as “clean up and try a different plan”

L C SL C S I need to ask a question of a systems wizard Plan 1: Locate A systems wizard in the E21 Monitor: check that person is still there Turn on the selected projector Monitor: check that projector turned on Project the message Done Monitor: check that the person noticed the message I don’t see light on the screen I see sally by the screen Projector-1 must be broken. We’ll try again, but using Projector-3. Plan Breakdown Recovering From Failures

L C SL C S Frobulate the sidetracker White Board Place Manager Sally is near The Whiteboard If A Person approaches A device ?x and Grammar ?y is relevant to ?x Then Activate Grammar ?y Activate the Drawing Grammar The Drawing Grammar Is Relevant to the Whiteboard Challenge 4: Response within Context The attention of the system should react to what people in the facility are doing. The system should choose its reactions to events based on the context –E.g. if the door opens, turn the lights on, but not if we’re watching a movie Perceptual interpretation should be biased by context –E.g. a person near the White Board, is likely to start drawing

L C SL C S Challenge 5: Perceptual Integration Separate the implementation of perceptual tasks from the uses to which perception is put Some modules advertise the class of “behavioral events” they are capable of recognizing and signaling –These events are organized into a taxonomy –The same event can be signaled by quite different perceptual modules (e.g. both face and voice recognition can localize a person). Other modules register their interest in certain classes of events –Requests at a higher level in the taxonomy subsume lower level events Modules which receive low-level events may register for and collate many different classes of events –They combine these and signal higher-level events Modules may request perceptual services when they are uncertain of their conclusions

L C SL C S Research Agenda: A Dynamic Event Bus For Perceptual Integration Visual Tracker Signal body motion Voice Identification Interested in body motion Signal the location of individuals Face Recognition Interested in face location I signal the location of individuals White Board Context Manager Interested in the location of individuals Signal people approaching the whiteboard Face Spotter Interested in body motion I signal face location Similar to “Blackboard” systems Publishers & Subscribers are “Knowledge Sources” Events are the blackboard data items But Highly distributed Use of Bayesian and Markov Techniques

L C SL C S Integration of Reactive and Goal Directed Processing Service Mapper Goals Plans Resource Allocator Actions Resource Pool Sensory Systems Events Blackboard Context Reactions Access Policies Plan Monitor Diagnosis & Recovery

L C SL C S Plan KB Goal Plan Knowledge Base Haystack s Knowledge acquisition Facilitato r Shared Plans Shared Info Shared Semantic Web Inference Resource Discovery Facilitator Agent Oxygen’s Knowledge-Based Infrastructure Helps People Collaborate: By acting as an assistant to meeting facilitators it can help to capture important information such as rationale By understanding the role of individuals within the organization it can help to decide who should see what information. By understanding the technical capabilities of individuals it can help to match “who can do what” to “what is required”. By understanding the context within which people act, it can act and react in appropriate ways. By interacting naturally and adaptively it can make knowledge capture “cost less than it’s worth”.