Header Biotechnologisches Zentrum

Research Projects at Technical University Dresden

 

Rule Responder Health Care and Life Science eScience Infrastructure

The goal of the project is a flexible and expandable distributed eScience rule inference infrastructure in the domain of Health Care and Life Science, which enables distributed deployment of rule responder service with dynamic integration and intelligent rule-based access to existing Web-based scientific services and data sources.

Type: Research Project
Position:

Project Lead

Status:

Ongoing

Technologies:

Rule Interchange Formats, Distributed Rule Inference Services, Complex Event Processing and rule-based Event Processing Languages, Multi Agent Systems, Semantic and Pragmatic Web technologies, Enterprise Service technologies and Enterprise Service Bus

Details: http://ibis.in.tum.de/projects/paw/hcls/
Contact: Adrian Paschke

Implementations within SourceForge Open Source and W3C HCLS

 

 

<< back

Research Projects at National Research Council Canada

Rule Responder: A RuleML-Based Distributed Pragmatic Agent Web and Enterprise Inference Service Architecture

 

Type: Research Project
Position:

Project Lead

Status:

Ongoing

Technologies:

Rule Interchange Formats, Distributed Rule Inference Services, Complex Event Processing and rule-based Event Processing Languages, Multi Agent Systems, Semantic and Pragmatic Web technologies, Enterprise Service technologies and Enterprise Service Bus

Details: http://responder.ruleml.org/
Contact: Adrian Paschke

Implementations within SourceForge Open Source

 

 

Research Projects at Technical University Munich

RBSLA: Rule-Based Service Level Agreements - Knowledge Representation for Automated e-Contract, SLA and Policy Management

Service-oriented architectures (SOAs) and Event-driven architectures (EDAs) provide a new paradigm for distributed computing. Automated management and monitoring of IT service contracts like Service Level Agreements (SLAs) or higher-level policies are essential for efficient and reliable service-oriented architectures (SOA) with high quality of service levels (QoS). The domain typically faces rapidly changing business and system environments, huge amounts of scattered contracts, business vocabularies and data, managed in distributed data sources, and a great variety of more or less cooperative roles involved during the contract life cylces. Moreover, correctness, reliability and traceability with respect to the contractual rules are vital to establish trust and fulfil legal compliance rules.

In this project we developed a rule-based knowledge representation (KR) to describe electronic contracts in a formal rule-based way. The research draws on logic programming (LP) techniques as well as on new standards in the area of web services computing (WSC) and the Semantic Web. A particular interest is the investigation of expressive logic programming techniques and logical formalisms such as defeasible logic, deontic logic, temporal event/action logics, transaction and update logics, description logics (external ontologies) as a means of deriving formal declarative contract specifications with which to reason about ideal and actual behaviours relating to agreed contract norms.

At the core of our contract management tool are rule-based languages to describe contracts like SLAs in a generic declarative way in a hybrid combination with external domain models such as Semantic Web-based business ontologies, relational data sources and object-oriented Java programs to integrate existing functionalities provided by.

The ContractLog KR and the superimposed RBSLA mark up language provide selected logical formalisms and inference services to adequately represent, manage, interchange and execute formalized rule-based SLAs or policy contracts. The ContractLog approach exploits logic programming techniques and combines them with Semantic Web technologies (RDFS / OWL ontologies) and procedural object-oriented Java and database functionalities in order to enable a tight integration of declarative rule specifications with existing external programs and systems such as Enterprise Java Beans, Web Services or System- and Networkmanagement tools or Semantic Web (business) vocabularies and relational or XML data bases which are typically used in the system, web and network management domain.

The RBSLA XML markup approach in contrast to existing languages in the SLA domain such as WSLA, WSOL, WS-Agreement, is not just a pure syntactical specification language with predefined language constructs, but is a declarative rule-based programming language with an operational and a declarative logic- based semantics for formalizing and implementing arbitrary contract-related functionalities and specifications. This declarative implementation-oriented design provides maximum flexibility and extensibility for SLA representation including the use of different external contract vocabularies written as Semantic Web ontologies such as WSMO, WS-Policy OWL ontology or KAoS and other ontologies such as OWL time, which can be integrated as external webized type systems into the logical SLA rules, giving them a domain-specific meaning.

Keywords: Service Level Management, Knowledge Representation and Knowledge Management, Contract Management, SLA Representation, Monitoring and Enforcement, Business Rules, Rules, Logic Programming, RDFS / OWL Semantic Web Ontologies, Rule-based Description Logic Programs, RuleML, RIF, Test-driven Verification and Validation

Type: Research Project
Position:

Project Lead

Status:

Ongoing

Technologies:

Extended Logic Programming, Semantic Web RDF/RDFS/OWL (Pellet, Jena), Java, Web Services, relational Databases (JDBC, SQL), XML (Datasources/Databases, RuleML, Reaction RuleML), JUnit, Test Coverage

Details: http://ibis.in.tum.de/projects/rbsla/index.php
Contact: Adrian Paschke

Implementations within SourceForge Open Source, RIF/RuleML and SemWebCentral

 

Reaction RuleML

Reaction RuleML is a general, practical, compact and user-friendly XML-serialized language for the family of reaction rules. It incorporates different kinds of production, action, reaction, and KR temporal/event/action logic rules into the native RuleML syntax using a system of step-wise extensions. In particular, the approach covers different kinds of reaction rules from various domains such as active-database ECA rules and triggers, forward-directed production rules, backward-reasoning temporal-KR event/action/process logics, event notification & messaging and active update, transition and transaction logics. It covers different execution styles for processing them such as active, where the reaction rules actively pull or detect the events possibly clocked by a monitoring/validity time function, passive, where the reaction rules passively wait (listen) on matching event instances, e.g. incoming event messages, which match with the define event definition patterns, and reasoning, where the focus is on the formalization of events and actions and reasoning on their effects on changeable knowledge states (fluents). Reaction rules can be specified on a global level in a tight combination with other rule types such as derivation rules or integrity constraints or locally, i.e. nested within other derivation or reaction rules. There are different evaluation styles for reaction rules such as strong and weak interpretation which are used to manage the "justification lifecycle" of local reaction rules in the derivation process of the outer rules. Reaction RuleML supports procedural calls on external procedures with side effects and enables expressive transactional OID-based updates on the extensional and intensional knowledge base, i.e. on facts and rules. Sophisticated postconditional verification, validation and integrity tests (V&V&I) using integrity constraints or test cases can be applied as post-conditional tests, which possibly might lead to roll-backs of the update actions. Complex events and actions can be expressed in terms of complex event / action algebra operators and different selection and consumption policies can be configured.

Keywords: Reaction Rules, Production Rules, ECA Rules, Active Rules, RuleML, ECA-RuleML, Event Logics, Action Logics, Event Calculus, Situation Calculus, Interval Calculus, Interval-based Event Calculus, Temporal Action Logic (TAL), Event Driven Architectures (EDA), Complex Event Algebra, Snoop, Transaction Logics, Update Logics, Dynamic Evolving Logic Programming, Production Rule Representation (PRR), Agent Communication, Agent Negotiation, FIPA ACL, Action Algebra, Event Condition Action ECA, Event Notification, Event Messaging,Rule Markup, RIF, Semantic Web

Type: Standardization Research Project (close relations to W3C RIF)
Position:

Co-Chair

Status:

Ongoing

Technologies:

XML, XMLS, XSLT, RuleML, EBNF, Semantic Web technologies

Details: http://ibis.in.tum.de/research/ReactionRuleML/

Contact: Adrian Paschke
Alexander Kozlenkov

Implementations within RuleML Initiative

 

Semantic-enriched, Ontology-based Content Management System

The ontology enhanced Content Management System supports unitized production and management of media contents (e.g. articles, pictures, films etc.). The new concept of construction plans based on different types of content modules (content-, layout-, semantic module) allows the reuse of once created contents for the production of different end products, e.g. online magazine, print magazine etc. At the core of the CMS are meta data standards for content descriptions and ontology-based repository structures in order to support the editors with powerful and comfortable search capabilities.

This project draws on Semantic Web standards such as RDFS and OWL to classify and search in digital content repositories using predefined ontologies. A Web-based prototype has been developed using the Jena API, Hybernate and the Lucene database. The software allows to annotage media contents such as pictures, films, structured and unstructured documents and navigate ontologies and query contents based on the ontology model.

Keywords: OWL, RDFS, Ontologies, Semantic Web, Digital Media

Type: Research Project in Cooperation with LMU
Position:

Project Lead

Status:

Finished

Technologies:

Java, Java Web Applications, Semantic Web technologies (RDFS, OWL Lite), Jena, Lucene, Hybernate

Details: http://ibis.in.tum.de/research/pastprojects.htm
Contact: Adrian Paschke Cooperation between TUM and LMU

 

 

Negotiation Pattern Language (NPL) and Design Pattern Library (DPL)

Currently negotiation covers a huge and unstructured domain of negotiation- and coordination protocols and associated strategies. Researchers and practitioners like system engineers have almost no support in discussing protocol solutions to frequently occurring negotiation problems and in designing and implementing successful protocols and associated negotiation media. Negotiation engineering remains a laborious trial and error process.

In this project we develope a novel approach to the description of negotiation and coordination protocols that is based on the idea of design patterns. Design patterns have been used successfully in recent years in the software engineering community in order to share knowledge about the structural and behavioural properties of software and to communicate object-oriented design solutions. We implement a descriptive, semi-formal design pattern language for negotiation and decentralized coordination protocol solutions, called Negotiation Pattern Language (NPL). NPL uses a combination of narrative human readable descriptions, formal models like Role Activity Diagrams (RAD) and machine-processable XML mark-up . This underpins tool-based cataloguing and publishing of patterns in a machine-readable, semi-formal pattern definition language and supports engineers to understand and apply these patterns.

Standard Semantic Web vocabularies (RDF based Dublin Core) are used to attach further meta data to the pattern definitions and the pattern documents are catalogued in a web-based Design Pattern Library (DPL) using OWL based taxonomy models (t-box model) to categorize the patterns as instances (a-box) of the ontologies’ topic classes. This allows meeting the following design objectives:

• A formal, machine-readable pattern definition language
• An open and extensible representation language facilitating shared terminologies
• A human-readable language with structured narrative and formal pattern descriptions
• Meta data annotations that allow reasoning about pattern descriptions, in particular to provide efficient search functionalities
• A format that fits nicely into the standard Web technologies and the upcoming Semantic Web

The online coordination and negotiation design pattern library contains an online glossary and a first collection of frequently used design patterns for the solution of decentralized coordination / negotiation problems (e.g. auctions, voting, dispute resolution, etc.). This should enable researchers and practitioners to communicate effectively about particular coordination protocols. The patterns are written from the point of view of a systems engineer, but draw on the results of other disciplines such as social choice theory, mechanism design theory or group decision and negotiation theory.

Our goal is to initiate a community effort and stimulate a constructive feedback from the electronic negotiation community in order to achieve more general consensus about protocol patterns, used terms and concepts and in the end provide a comprehensive collection of negotiation design patterns.

Keywords: Design Patterns, Negotiation and Coordination Protocols, Agent Negotiation, Voting, Auctions, Coalition Forming, Bilateral Negotiation / Bargaining and Markets, Contracting, Dispute Resolution, Distributed Problem Solving, Heterogenous Information Integration, Semantic Web Technologies, Meta Data Annotation

Type: Research Project
Position:

Project Lead

Status:

Finished

Technologies:

Java, Java Web Applications (JSP), Semantic Web technologies (Dublin Core, OWL Lite), Jena, native XML database (Tamino), XQuery, Google Web Service API

Details: Paschke, A., Kiss, C., Al-Hunaty, S.: NPL: Negotiation Pattern Language - A Design Pattern Language for Decentralized (Agent) Coordination and Negotiation Protocols, In: Banda, R. (Ed.), In: E-Negotiation - An Introduction, ICFAI University Press, ISBN 81-314-0448-X, September 2006. (pdf)
Contact: Adrian Paschke supported by the Canadian INE SSHRC E-negotiation research program

 

Prova - Rule-based Java Scripting for Distributed Web Applications and the Semantic Web

Prova positions itself as a rule-based scripting (RBS) system for web-based middleware. The language breaks new ground in combining imperative object-oriented and declarative programming.

Prova is derived from Mandarax Java-based inference system developed by Jens Dietrich. Prova extends Mandarax by providing a proper language syntax, semantics for extended LPs, native syntax integration with Java, and agent messaging and reaction rules.

The design goals of Prova

  • Marry the benefits of declarative and object-oriented programming;
  • Combine the syntaxes of Prolog and Java -ultimate logic and object-oriented languages;
  • Expose logic and agent behaviour as rules;
  • Access data sources via wrappers written in Java or command-line shells like Perl;
  • Make all Java API from available packages directly accessible from rules;
  • Run within the Java runtime;
  • Enable rapid prototyping of applications;
  • Offer a rule-based platform for distributed agent programming and reactive Event-driven Architectures (EDA) 

Keywords: Hybrid Backward Reasoning Rule Engine, Meta-data annotaded, multi-sorted, prioritized, modular , scoped, transactional extended logic programming with order-sorted module based scripting syntax, backward-reasoning with linear goal memoization (tabling) and extended key indexing for configurable semantics variants

Type: Research Project
Position:

Project Lead together with Alexander Kozlenkov (Betfair) and Michael Schroeder (Biotec, Dresden)

Status:

Ongoing

Technologies:

Java, Mandarax, Semantic Web (RDF,RDFS,OWL), Jena, Pellet, RuleML, Reaction RuleML, JDBC/SQL, JMS, Jade, Mule, Web Services, Maven

Details: Prova
Contact: Alexander Kozlenkov, Adrian Paschke, Michael Schroeder Implementations within Mandarax SourceForge Open Source

 

Mandarax RDF/OWL and Event Calculus Extension

In this project several extensions to the Mandarax Java-based inference system for using Semantic Web RDF, RDFS and OWL, expressive logical formalisms such as the Event Calculus have been developed.

Keywords: Java-based backward-reasoning rule engine,RDF/RDFS/OWL extension, Event Calculus

Type: Research Project
Position:

Chief Developer

Status:

Finished

Technologies:

Java, Mandarax, Semantic Web (RDF,RDFS,OWL), RuleML

Details: http://mandarax.blogspot.com
Contact: Adrian Paschke, Jens Dietrich Implementations within Mandarax SourceForge Open Source

Projects at Friedrich Alexander University (FAU) Erlangen/Nuernberg

 

PAMAS –AN AGENT-BASED SUPPLY CHAIN EVENT MANAGEMENT SYSTEM

An agent-based supply chain event management and tracking system using forward-reasoning production rules (Jess), multi-agent technologies (FIPA-OS and FIPA-ACL) and heuristics to monitor, (pro-)actively detect and quantify incidents and disruptions in supply chains.

Keywords: Supply Chain Management, Supply Chain Event Management and Tracking, Production Rules (Jess), Multi-Agent Systems (MAS), FIPA

Type: Industrial Research Project. Cooperation between Siemens AG and the Friedrich Alexander University (FAU).
Position:

PAMAS Project Lead

Status:

Finished

Technologies:

Java, Jess, FIPA-OS, FIPA-ACL, ERP SAP (SAP BAPI)

Details:
  • Paschke, A., Zimmermann W.: A Multi Agent System for Supply Chain Tracking and Event Management in Journal of International Technologies and Informations Management (JITIM), 12/03. 
  • Paschke, A., Zimmermann W.: PAMAS – An Agent-Based Supply Chain Event Management System, paper Americas Conference on Information Systems (AMCIS03), August 2003.  (pdf)
Contact: Adrian Paschke Funded by Siemens AG

 

3D Virtual Reality Catalogs

The project exploits virtual reality technologies (Java3D and VRML) to present 3D electronic catalogs on the internet.

1. Subproject was in cooperation with the industry partners Adidas (sports)

The goal of this subproject was to vizualize products such as sport shoes, shipping boxes etc. in a web-based 3D editor, in oder to choose attributes such as color and place logos, signs or text on them. This should allow customers to design their individualized products, which can then be produced according to the customer wishes. The 3D technology gives the customer a better impression of the item, since he can roll the object, take a look from all sides or zoom in or out.

2. Subproject was in cooperation with the industry partners BIK

In this project a 3D catalog of building materials for the do-it-yourselfer has been developed. The customer can upload 3D models of its house or appartment or use predefined models. The select materials such as wallpapers, planks or paint can be applied to the virtual house and the customer can walk through the complete 3D virtual world. He can get a much better impressenion of how the material will look in a real world and he can change the design by selecting other materials from the catalog.

A particular gimmick of the 3D catalog is that the program also computes the needed number of materials according to the dimension of the object in the 3D world, e.g., the floor of a room, or all walls but without the windows and doors. This computation needs to consider the package size of the product (typically more than one item is bundled and sold together) and the clipping which is dependend on the product. This additional information is stored as meta data together with the catalogs' products.

Keywords: 3D Virtual Electronic Catalogs, Virtual Reality, Individual Differentiation of Variants

Type: Industrial Research Project with Adidas and BIK.
Position:

Project Lead

Status:

Finished

Technologies:

Java, Web Application Technologies, Linear Programming, Java3D, VRML, Cosmo

Details:

Sorry - IP protection

If you want more information please contact me.

Contact: Adrian Paschke Funded by Adidas, BIK


footer line  © Biotechnologisches Zentrum der TU Dresden | Impressum