Keynote:
David Embley, Research Challenges in Active Conceptual Modeling of Learning

Paper 1
Ajantha Dahanayake and Bernhard Thalheim, Towards a Framework for Emergent Modeling

Paper 2
Faiz Currim and Sudha Ram, When Entities are Types: Effectively Modeling Type-Instantiation Relationships

Workshop Chairs:
Hannu Kangassalo - University of Tampere, Finland.
Salvatore T. March - Vanderbilt University, USA.
Leah Y. Wong - SPAWARSYSCEN Pacific, USA.

Problem:The advent of information technology allows us to model the world by mapping real-world scenarios onto information systems and applications in a more sophisticated way.  However, today?s databases and knowledge bases only reflect the static characteristics of the intended Universe of Discourse, captured by the conceptual model as distinct snapshots.  The information system, which provides us with ?almost recent? information, neither supports applications that require historical information nor provides information for projecting the future based on past experience and lessons learned. Without the relationships between snapshots, it is difficult to simulate the ?what if? scenarios.  Temporal and spatial relationships between entity behaviors and uncertainty cannot be fully modeled. Temporal concepts are not taken into account properly. Therefore, historical information and their changes cannot be managed, and the certainty of information cannot be assessed. Inadequate dynamic modeling constructs (e.g. perspective, dynamic relationships and changes, degree of importance of relationships) result in incomplete representation of the changing real-world domain.

Approach: To achieve active information processing, learning from our past experience is essential. Learning is a continuous process by which relatively permanent behavioral changes occur, potentially as a result of an experience. Lessons learned are knowledge gained by reflecting on experiences that can avoid the repetitions of past mishaps to share observations and to improve future actions. While learning is an ongoing process that transfers knowledge from one state to another, a lesson learned summarizes knowledge at a point in time. To describe an experience is to model past events and associated knowledge from a different perspective. This historical perspective allows us to describe a lesson learned from the interaction of episodic and semantic memories. The domain can be described in terms of topic, time/space, people, scenarios/events, cause/effect and general knowledge about the situation or domain. Active conceptual modeling is a continual process of describing all concepts and aspects of a domain, its activities, and changes under different perspectives. The model is viewed as a multilevel (e.g. strategic, tactical, operational) and multi-perspective high-level abstraction of reality. Our effort focuses on relationships between past knowledge/data and current knowledge/data from different perspectives. We propose a framework for active conceptual modeling of learning.

Conventional conceptual modeling for database design is a simple case of active modeling. The active conceptual model will provide the necessary control and traceability for the evolving domain.  The moving snapshots could potentially become frames for creating a movie of the past, present, and future; and help simulate the target systems to answer the ?what if? questions.  It will also allow us to continually learn and make inferences to provide foresight from hindsight. The user will be notified when the alerted situations are detected, based on the monitoring requirements under dynamic constraints and related information from the underlying data sources and learning-base.

Topics:

Technical Areas: Accomplishing our goal will require investigation of the following basic and exploratory research areas. Some other relevant areas may also be found.
• Integrating time, space, and perspective dimensions in a theoretical framework of conceptual models
     - Theory of human concepts, human cognition
     - ER theory
     - Mathematical active conceptual models
     - Multi-level conceptual modeling
     - Multi-perspective conceptual modeling
     - Multi-media information modeling
     - Mapping of constructs among conceptual models

• Management of continuous changes and learning
     - Conceptual change
     - Continuous knowledge acquisition
     - Experience modeling and management
     - Learning from experience
     - Representation and management of changes
     - Transfer learning in time dimension
     - Lessons learned capturing
     - Information extraction, discovery, and summarization

• Behaviors of evolving systems – including model evolution, patterns, interpretation, uncertainty, integration
     - Time and events in evolving systems
     - Situation monitoring (system- and user-level) 
     - Schema evolution and version management
     - Content awareness and context awareness
     - Modeling of context changes
     - Information integration and interpretation
     - Pattern recognition over a time period
     - Uncertainty management WRT integrity
     - Reactive, proactive, adaptive, deductive capability in support of active behavior
     - Combined episodic and semantic memory paradigm for structuring of historical information

• Executable conceptual models for implementation of active systems
     - Dynamic reserve modeling
     - Storage management
     - Security 
     - User interface
     - Bench marking for Test & Evaluation
     - Languages for information manipulation
     - Architectures for information system based on the active conceptual model

Capability: The active model can only be realized by integrating technology (e.g. AI, software engineering, information/knowledge management, cognitive science, philosophy, etc.) and combining modeling techniques. We will provide an enhanced situational awareness and monitoring capability through the following services:
•    Information provenance for understanding and interpreting information in a holistic manner
•    Contextual information integration with uncertainty indication
•    Rewind memory to specific times and/or situations and move forward with different assumptions
•    Trace scenarios and discover hidden and implicit relationships between events
•    Detect changes in evolving situations over time
•    Situation monitoring with context-awareness adaptation
•    Event patterns discovery (related and seemingly unrelated events)
•    Anomaly detection with respect to situational changes
•    Notification when similar situations are occurring
•    Dynamic reserve modeling to derive context models based on data at a given time 
•    Learning, inferring, and reasoning of forensics and present for predicting future actions
•    Interactive user interface

Applications: The ACM-L capability can be applied to a large class of applications including the following:
•    Active learning
•    Adaptive C4ISR 
•    Counter-terrorism for tracking “surprises”
•    Homeland Security   
•    Info-Forensics
•    Law enforcement
•    Lessons-learned systems
•    Medical/patient information systems
•    Situation awareness and monitoring (MDA, GWOT)
•    Simulation and modeling
•    Many others…

Status: To begin framing the problem, SPAWARSYSCEN Pacific hosted two workshops on ACM-L in 2006. The first event was held at SPAWARSYSCEN Pacific to introduce the Science & Technology (S&T) Initiative and identify a Research and Development agenda for the technology development investigation. Eleven invited experts in Conceptual Modeling presented position papers on the proposed S&T Initiative. The first open workshop was held at the 25th International Conference on Conceptual Modeling, ER 2006, 6-9 November 2006, in Tucson, Arizona. An application, AWARE (Active Wisdom Advancing Retrospective Exploration), is being identified. The second open workshop was held at the 28th International Conference on Conceptual Modeling, ER 2009, 9-12 November 2009, in Gramado, Brazil. The workshop covered a wide spectrum of issues from operational requirements to basic research in concepts, learning, thinking, and communication.

Workshop deadlines:

Abstract Submission:	April 20, 2010
Full Paper Submission:	April 28, 2010
Author Notification:	June 7, 2010
Camera-ready Paper Submission:	June 30, 2010
Workshop:	November 1-4, 2010

Paper Submission

Formatting Guidelines

ACM-L 2010 proceedings will be part of the ER 2010 Workshop volume published by Springer-Verlag in the LNCS series. Thus, authors must submit manuscripts using the Springer-Verlag LNCS style for Lecture Notes in Computer Science. Refer to http://www.springer.de/comp/lncs/authors.html for style files and details. Papers in the final proceedings are strictly limited to 10 pages. Therefore, submitted papers should also not exceed 10 pages, but technical appendices, e.g. containing proofs, can be added to a submission.
 Papers must be in English, formatted in LNCS style and submitted as PDF-files. Submitted papers must be original and not submitted or accepted for publication in any other workshop, conference, or journal.

Submission Guidelines

Submission to ACM-L 2010 will be by electronic mail, only, to all three workshop chairs to addresses below. Submission must be in PostScript or PDF format, by the due date. All correspondence with authors will be via e-mail. Please ensure that your submission includes an e-mail address for the corresponding author.

Workshop chairs and their e-mail addresses:

          Hannu Kangassalo;  University of Tampere, Finland; hk@cs.uta.fi
          Salvatore T. March;  Vanderbilt University, U.S.A; Sal.March@owen.vanderbilt.edu
          Leah Y Wong;  SPAWARSYSCEN Pacific, U.S.A; leah.wong@navy.mil

PROGRAM COMMITTEE MEMBERS (to be extended)

Stefano Borgo, Laboratory for Applied Ontology, ISTC-CNR, Italy
Alfredo Cuzzocrea, University of Calabria, Italy
Giancarlo Guizzardi, Universidade Federal do Espírito Santo, Brazil
Raymond A Liuzzi, Raymond Technologies, USA
Jari Palomäki, Tampere University of Technology/Pori, Finland
Oscar Pastor, Valencia University of Technology, Spain
Sudha Ram, University of Arizona, USA
Laura Spinsanti, LBD lab ? EPFL, Swizerland
Il-Yeol Song, Drexel University, USA
Bernhard Thalheim, Christian Albrechts University Kiel, Germany