Example Completed
Projects
(Projects
for 2007 listed here)
Placement details
The HP Labs Bristol
Semantic Web Group regularly host students. Some of these are
undergraduates undertaking an industrial placement as part of
their degree, some are carrying out the research portion of
their Masters thesis based in labs, others are taking a
sabbatical during their PhD. Placements last between 4 and 12
months.
Precise start and end dates will depend on the regulations of
each student's university, but we are expecting each student's
duration of stay to be six months, typically starting in
May.
We are looking for students (at any level) who are already
actively involved with Semantic Web research and development,
for example with one of a published paper or poster, a
contribution to a software development project (ideally open
source), active involvement in a Semantic Web mailing list, or
other contribution to the field. The placements would also suit
MSc students who are required to do a research project and
dissertation, or people studying for a PhD who are able to take
a sabbatical.
We have a number of
projects we are proposing, though we are also open to
students proposing their own. Interested applicants should
contact the semantic web group at semweb@hpl.hp.com with their
resumé, and a covering letter indicating which project(s)
they are interested in and why they think they would be a good
candidate for working with the group in that area. Candidates
with their own project proposal should send details together
with why they think it is relevant to the research agenda of the
group (see http://www.hpl.hp.com/semweb/hpl-research.htm
).
After an initial screening on resumés, we will phone
screen a shortlist of candidates, and where possible follow up
with interviews at the HPLB site. This will be an opportunity to
look around and to meet the Semantic Web team to discuss
specific projects. HP will pay your travel expenses.
Hewlett-Packard Labs Bristol (HPLB) is the main European
center for HP's worldwide research labs. HPLB has a long and
successful tradition of hosting students, from various
universities in Europe and beyond. Students spend the duration
of their research project (including the writing of their
thesis/dissertation) at HPLB, are given full access to HPLB
facilities, and are paid a standard monthly rate for the
duration of their stay. Each student is assigned a member of
HPLB staff to act as a day-to-day supervisor but could also be
in regular supervisory contact with a member of faculty at their
university. The student's dissertation may be published as an HP
Labs technical report, and it is also hoped (but not required)
that the student's research work will lead to a publication in a
suitable peer-reviewed international conference or journal, and
potentially also to filing of one or more international patent
applications.
HPLB has established procedures for helping MSc students with
relocation. Non-EU students may need to check the conditions of
their UK visas, but HP is a global company and staff at HPLB
have experience of dealing with students from many countries.
Hewlett-Packard is fully committed to non-discrimination and
equal opportunities.
(top)
BrownSauce RDF Browser
There is RDF data all over the place, in XML documents and
sources like databases. Getting all that data is impractical,
and would be unreadable (to say the least). So BrownSauce is an
attempt to make something which can browse that information.
BrownSauce breaks the problem into two parts: coarse-graining
(breaking the data down into usable chunks, like 'information
about person X) and aggregation (making those chunks from
multiple sources). The first part is done, and users can browse
more than one source using rdfs:seeAlso references.
Aggregation is currently being worked on.
BrownSauce runs as a local http server, or can be added to a
java web application server like Tomcat or Jetty. The current
interface is HTML and can be styled using CSS (the HTML is
marked up using classes relating to the RDF). Other interfaces
should be simple to implement.
BrownSauce was written by Damian Steer whilst employed at
HP Labs Bristol. BrownSauce is free
software, released under a
BSD style licence. It can be downloaded from SourceForge at
http://sourceforge.net/projects/brownsauce/.
HP
Labs Technical Report HPL-2003-10
(top)
The goal of this project was to explore using ontologies to
improve free text. The aim was to enable users to search not
only for particular terms, but also for semantically related
terms. For example given a medical ontology, it is possible to
search for references to cancers, or given FOAF data it is
possible to search an email archive for friends of a person.
The implementation was based on Lucene (the Open Source free
text search engine from the Apache foundation). The Lucene query
language was extended to enable users to express searches in
terms semantic operators on terms. The design allowed easy
addition of new semantic operators which were implemented as
RDQL queries on an ontology. A demonstration supports searching
for synonyms, hyponyms, hypernyms and instances of terms using a
translation of the WordNet lexicon to OWL.
This work was done during 2004 by Viral Parekh who was on
placement at HP Labs Bristol.
(top)
Agent front end to a semantic web service
This work involved the development of an agent-based system,
using the NUIN framework, as an experimental front-end for the
HP Software Marketplace. This is a business critical B2B
web-service deployed by the HP European Software Centre in
Galway, Ireland. The aim was to develop tools for syntax
independent interaction by lifting XML message data into, and
out of, a common representational framework, namely RDF. The
results of this work were successfully applied in European
project: Semantic Web enabled Web Services (IST-2002-37134). The
work was presented at
the first AKT workshop on Semantic Web Services
This work was carried out by Yathiraj Bhat Udupi of North
Carolina State University, Raleigh NC USA. The work was
supervised by Steve Battle of HP Labs Bristol
(top)
This work examined relationships between service oriented
architectures and the semantic web. In particular this focused
on the use of service ontologies for use by web-agents. The
objective was to take descriptions in the OWL-Service ontology
(OWL-S) and compile them into executable form. The target
language was NUIN script, a language for expressing agent plans
in a Belief/Desire/Intent framework. The results of this work
were successfully applied in European project: Semantic Web
enabled Web Services (IST-2002-37134).The work was presented at
the first
AKT workshop on Semantic Web Services.
This work was carried out by Renato de Freitas Bulcão
Neto of the University of Sao Paulo, Brazil. The work was
supervised by Steve Battle of HP Labs Bristol
(top)
Snippet
Manager
Paolo Castagna worked with us during 2005, drawing on our
semantic blogging and semantic portal work to create a snippet
manager. The objective was to create a lightweight method of
capturing and organising a set of information items for use by a
small peer group. We created a prototype application that we
demonstrated at ISWC 2005.
(top)
|
printable version
