Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-discussion] JOB: Artificial Intelligence Software Developer, Washington DC Area

Debbie Duong debbieduong62 at gmail.com
Tue Feb 10 11:30:07 PST 2009


Please send reply to <debbieduong62 at gmail.com>

>
>
> *Job Category:* Artificial Intelligence Software Developer
>
>
>
> *Clearance:* US DoD Secret required.
>
>
>
> *Experience:* Five years experience in Artificial Intelligence related
> programming or a Master's Degree in Artificial Intelligence or a related
> field.
>
>
>
> *Skills:* Experience in the following, or similar, subjects a plus:  Software
> Patterns, Agent-Based Simulation, Software Optimization and Scalability,
> Open Source Contributions, Ontologies, Inference Engines, Evolutionary
> Computation, Neural Networks, Bayesian Networks, Fuzzy Expert Systems, Data
> Mining, Case-Based Reasoning, Game Trees and Game Theory, Social Network
> Analysis, Statistical Design of Experiments.
>
> * *
>
> *Programming Language:*  Java
>
> * *
>
> *Other Software and Development Tools:*  Experience in the following, or
> similar, software a plus - Protégé, Owl, Pellet, Jena, Jastor, Weka, Repast,
> Groovy, ECJ, Ptolemy, JFuzzyLogic, Joone, Jung, R, Jboss, Prefuse
>
>
>
> *Job Description:*
>
>
>
> Artificial Intelligence Software Developers needed for cutting-edge
> Computational Social Science simulation project.   Will be working in a
> Team based environment with other SW developers.
>
>
>
> Potential Task areas include:
>
>
>
> *i)        **Scientific Experimentation and Data Mining:*  Enhance
> existing software to support finding patterns in wargame events, for the
> purposes of testing hypotheses about relations between events and for
> discovering relationships between events.   Incorporate open source data
> mining and artificial intelligence software, such as, for example, Weka and
> ECJ, to automatically find patterns in moves and outcomes
>
>
>
> *ii)      **Interoperation of Hybrid Models:*  Support the interoperation
> of hybrid models, to ensure that models that have multiple resolutions and
> perspectives share meaning.   Through xml, implement the translations
> between models.  Integrate with integration tools that support semantic
> interoperation through ontologies, and other methodologies such as, for
> example, the COMPOEX backplane or Ptolemy.   **
>
> * *
>
> Help implement a Hub and Spoke design for translation between data models a
> system by which simulation models and data of different data models may
> interoperate through their own individual data models, a common data model,
> and a translation data model between their own individual data models and
> the common data model.
>
>
>
> Enhance the software enforcement of a data model "contract" using
> ontology-based software engineering techniques.  Extend an existing
> example implemented in Jena and Jastor, converting it to a Dynamic Object
> Model (DOM) language (such as, for example, Groovy).
>
>
>
> *iii)    **Support for Conflict Resolution between models.  *Support the
> conflict resolution of possibly conflicting hybrid models, for the purposes
> of making a unified, coherent picture of the social environment.
>
>
>
> Integrate open source software (such as, for example, JFuzzyLogic) to match
> the simulation output to correlative social study data, and establish the
> correlative relations that should exist between and within component social
> science models in support of validation and consensus building of possibly
> conflicting models.
>
>
>
> Implement a framework for consensus-building.  Make possible the
> specification of arbitrary schemes for developing a model consensus.  The
> framework would have a way to handle issues of integration, for example, the
> models may conflict by having mutually exclusive or uncorrelated behaviors.
> The framework would allow various consensus schemes to be switched in and
> out.  Implement a simple prototype weighted voting scheme in the framework
>
>
>
> *iv)    **Support for Component Models*.  Enhance the Nexus Intelligent
> Adaptive Agent Based Models to increase their generality, scalability,
> efficiency, and ability to work as component models for other software.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/protege-discussion/attachments/20090210/e5e189c3/attachment.html>


More information about the protege-discussion mailing list