Knowledge representation and Semantic Web

The ongoing goal of the Semantic Web is to allow computers to analyse and manipulate information in more sophisticated ways through the use of appropriate knowledge representation and reasoning mechanisms. Our research covers fundamental issues arising in logics and theories of cognition as well problems arising in practical systems. We seek to improve human-computer interaction through the development and evaluation of appropriate knowledge representation and reasoning techniques for specific contexts. Current research includes investigation of logics for reasoning with spatial and temporal information, use of ontologies for providing users with more effective ways of searching for information, representing and using knowledge about learners and learning objects, and representing and querying community knowledge.

Meaningful mining and visualisation of data from RSS feeds
Friday, 20 August 2010

oshea_martin.jpg

 First supervisor

Prof Mark Levene

This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

 

Second supervisor

Dr Dell Zhang

This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

Research details

Work started in late 2008 and is expected to finish in late 2011 / early 2012.

 

Keywords

social data

RSS feeds

visualisations

dataset

social networking

 

Key themes

The first stage of my PhD work began with the investigation of on-line data-centric social networking. This work saw the development of myDataSharer (myDS) which aimed to unite data and community into an application which might have potential real-world uses by allowing users to upload and visualise various forms of data. A usage experiment of myDS's alpha version was successfully completed in late 2009. From the results of this experiment, my work is now concentrated on meaningfully mining and visualising data from raw RSS feeds. The next stage of my work will be a Web 2.0 application dedicated to this end to be released onto the internet late in 2010. My thesis in 2011 will then document the results of this work within the context of social networking and social data analysis.

 Results to date

In late 2009 for the Search Engines and Web Navigation module of the Birkbeck DCSIS MSc courses, under Prof Levene's supervision, myDS was used as the platform for an assessed coursework experiment by 35 students to mine and visualise meaningful data from RSS feeds. The results were successful with some 173 datasets created along with an equal number of visualisations. This has provided me with a great deal of mined data and usage metadata for analysis and the aim is to publish these results later in 2010.

 Why the Knowledge Lab?

Birkbeck DCSIS and IOE staff at the London Knowledge Lab have been able to advise and help me with many specific aspects of my work and PhD generally over the last two years which has been of benefit. Also, LKL is very conveniently placed in central London near to Birkbeck, to allow for presentations or demonstration which I have been involved in or for supervising student project.

oshea_martin_diag1.gif

myDataSharer's concept.

 Key publications (if any) here

Mining and visualising meaningful data from RSS feeds: a case study, Martin O'Shea and Mark Levene, 2010. (To be published during 2010).