Lehigh University

SWAT Projects

  SWAT Home
  People
  Projects
  Publications
  Downloads
  Resources
  Contact Info

Current Projects

Inactive Projects



Hawkeye: Towards Scalable Integration of Semantic Web Sources
Description:
At present the Semantic Web consists of numerous independent ontologies. We propose that OWL can be used to integrate these ontologies and thereby integrate the data sources that commit to them. By using OWL as opposed to a new integration mechanism, it allows us to leverage from a suite of existing well tested tools and techniques. Hawkeye is a system that given sufficient OWL descriptions, can answer queries that span heterogeneous data sources. Our system is influenced by the contemporary search engine model of querying a centralized repository. Although this can lead to staleness of answers, it can also lead to significant performance improvements. In order to perform scalable reasoning, Hawkeye loosely couples a description logic reasoner with a relational database management system.

This project will load as much real world Semantic Web data as possible into a single knowledge base and demonstrate realistic integration queries for various scenarios.

Selected Publications:
"An Investigation into the Feasibility of the Semantic Web"
Contact:
Zhengxiang Pan
Funding Sources:
NSF1

Back to project list


Knowledge Base for DARPA DTN
Description:
This project will adapt DLDB approach into the Disruption Tolerant Network scenario. The goal is to build a persistent knowledge base (KB) that offers a flexible and extensible framework for:
  • derivation of facts from simpler facts by applying inference rules
  • uniform query interface for both explicitly-stated and derived facts

This project will also provide a benchmark solution to evaluate the performance and capabilities of DTN data stores.

Contact:
Zhengxiang Pan
Funding Sources:
DARPA (via subcontract from BBN)

Back to project list


Tools for Ontology Versioning
Description:
This project will build a series of tools for ontology versioning by using the outcomes from the project of Semantics for Distributed Ontologies. We have built Virtual Perspective Interface (VPI), which adds support for dynamic ontologies to existing knowledge bases. We currently plan to implement another tool that can determine the semantics and syntactic differences between two ontologies.
Downloads:
See our download page to obtain the latest build of VPI.
Selected Publications:
"Adding Support for Dynamic Ontologies to Existing Knowledge Bases"
Contact:
Zhengxiang Pan
Funding Sources:
DARPA (via subcontract from ISX)

Back to project list


Semantics for Distributed Ontologies
Description:
The knowledge of the Semantic Web will be the products of millions of autonomous parties and may represent many different viewpoints. However, the Semantic Web lacks a strong underlying theory that considers its distributed aspects. Although languages such as RDF and OWL currently give a definitive account of the meaning of single document, things become more ambiguous when you consider how documents should be combined.

This project is developing model theoretic and modal logic semantics to describe distirbuted ontologies, and in particular to account to ontology versioning.

Selected Publications:
"A Model Theoretic Semantics for Ontology Versioning"
Contact:
Zhengxiang Pan
Funding Sources:
NSF1

Back to project list


HAWK: OWL Repository and Toolkit
Description:
HAWK is a repository framework and toolkit supporting the latest web ontology language -- OWL. HAWK project includes the OWL parser, memory based and database based persistent models, query API, reasoning interface, etc. HAWK is currently under development, new features are being added through its open architecture.
Downloads:
See our download page to obtain the latest build.
Contact:
Zhengxiang Pan
Funding Sources:
DARPA (via subcontract from Lockheed Martin)2

Back to project list


Comparison of Semantic Web Repositories
Description:

We are currently conducting experiments to determine the tradeoff space for various implementations of OWL repositories. A number of systems will be considered, including relational databases, simple knowledge representation systems, deductive databases, and description logic systems. Clearly, there is a tradeoff between logical completeness (with respect to the full language) and performance of the system. The goal of this research is to determine the degree of the tradeoffs, which will help users better choose repositories that meet their specific needs.

As part of this project, we have developed the Lehigh University Benchmark (LUBM). This benchmark facilitates the evaluation of Semantic Web repositories. In particular, it evaluates the performance of those repositories with respect to extensional queries over a large data set that commits to a single realistic ontology. In addition to measuring performance, the benchmark also measures the degree to which a system is sound and complete on a range of queries.

Downloads:
See LUBM homepage for more information and downloads.
Selected Publications:
See LUBM homepage for related publications.
Contact:
Yuanbo Guo
Funding Sources:
NSF1, DARPA (via subcontract from Lockheed Martin)2

Back to project list


DLDB

Note:DLDB is now part of the HAWK project. Please see HAWK for more information.

Description:
A knowledge base system that extends a relational database management system with additional capabilities for DAML+OIL inference. We have implemented a lightweight version that makes use of a common RDBMS (MS Access) and the FaCT description logic reasoner, that is expected to be adequate for personal or small-business usage.
Downloads:
See our download page to obtain the latest build.
Selected Publications:
"DLDB: Extending Relational Databases to Support Semantic Web Queries"
Contact:
Zhengxiang Pan
Funding Sources:
NSF1, DARPA (via subcontract from Lockheed Martin)2

Back to project list


Semantic Web Information Integration
Description:
This work builds on research in information integration to develop tools that allow Semantic Web resources to be queried and integrated in real time. Although information integration techniques have seen success in the database area, they are difficult to apply in a web environment that is open, autonomous, and lacks the structure of a database. The Semantic Web extends the existing web with structure, and a mechanism to specify formal and shareable semantics. Therefore the Semantic Web may be a better environment to apply information integration techniques. Our work aims to formalize an information integration framework based on Semantic Web technologies, define a subset of OWL that is compatible with the state of information integration formalisms and develop a novel distributed query answering algorithm.
Selected Publications:
Information Integration via an End-to-End Distributed Semantic Web System.
Contact:
Abir Qasem
Funding Sources:
DOE (via subcontract from Tech-X)

Back to project list


OntoPlan
Description:
In conjunction with Dr. Hector Munoz-Avila, we have looked at integrating Semantic Web technology with an HTN planner in order to perform web service composition with heterogeneous sources. A recent focus has been on augmenting Semantic Web languages with local closed world (LCW) capabilities. The nature of the Semantic Web means that it must be inherently an open world, you cannot assume something is false simply because you have not seen it asserted. However, having to explicitly express all false information can be particularly cumbersome. LCW statements, originally introduced by Etzioni, Golden, and Weld, have been proposed to solve this problem in both information integration and planning. We are exploring how to adapt this work to the Semantic Web.


Selected Publications:
LCW-Based Agent Planning for the Semantic Web
Contact:
Abir Qasem

Back to project list


Querying the Inconsistent and Untrustworthy Web
Description:
This work investigates on how to obtain useful information in an environment like the Semantic Web, which is assumed to be inconsistent and untrustworthy. We look into the issue from the perspective of extensional queries on the Semantic Web. Specifically, we attempt to invent mechanisms that could efficiently answer queries about: if a set of trusted document is consistent and entails a specific statement; and which set of documents has to be trusted in order for a specific statement to be true.
Selected Publications:
This paper discusses our intial proposal: "An Initial Investigation into Querying an Inconsistent and Untrustworthy Web".
Contact:
Yuanbo Guo

Back to project list


MS Access Repository for SHOE
Description:
This repository allows SHOE information to be stored in a MS Access database. We have modified the standard SHOE tools, including Expose and SHOE Search to work with this repository, making a complete SHOE system available to Windows users.
Downloads:
See our download page to obtain the source or application files.
Contact:
Zhengxiang Pan

Back to project list


Footnotes:

1 - This material is based upon work supported in part by the National Science Foundation under Grant No. 0346963. Any opinions findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF).

2 - This project has concluded.