ExaMiner

The following people are involved with this project:

More about this project

The ExaMiner project is a collaboration between the R&D/ILRT and the Intelligent Systems Laboratory at the University of Bristol. The work is part of the University's Exabyte Informatics research theme which aims to advance computing as the language of 21st century science. The project builds upon two recent exploratory Exabyte Informatics applications:

1. Exabyte Research Hub The Exabyte team has designed and implemented a first prototype of the Exabyte Research Hub, which is a computational platform allowing scientists to store, curate and share their data; explore, apply and share computational methods enhancing the data; and pursue their aims as part of a larger community. The platform was demonstrated live in October '09 at the HPC Symposium, and has already led to several possible avenues of interdisciplinary collaboration.
2. SubSift The R&D/ILRT and the Exabyte team collaborated in the JISC-supported SubSift project to support academic peer review. SubSift matches submitted conference or journal papers to potential peer reviewers based on the similarity between the paper’s abstract and the reviewer’s publications as found in online bibliographic databases such as Google Scholar. The project was recently delivered in the form of web services designed to support not only peer review but also personalised data discovery and mashups, and has already been used by several major data mining conferences.

The ExaMiner project will apply and further enhance the above tools to deliver functionality for mining and mapping the UoB research landscape. Specifically, the ExaMiner software will

1. generate automatic profiles of researchers and research groups from UoB web pages;
2. allow visual navigation of these profiles, as well as navigation by free-text query;
3. provide social networking components, including the ability to store a personal profile and a network of interesting contacts, events and connections;
4. notify a user of new resources in their sphere of interest; and
5. improve performance over time by responding to user feedback.

It is important to note that this is neither a pure research project nor pure IT development but a combination of both, and as such complementary to related projects at the University. Most closely related is R&D/ILRT's ResearchRevealed project, which is primarily concerned with applying Semantic Web technology to create a single integrated view and query point for a broad range of otherwise unconnected repositories within the University. The main differences are that ExaMiner mines free-form web content rather than institutional repositories and applies state-of-the-art Artificial Intelligence techniques to uncover hidden connections.

Possible users include academics, research managers, current and prospective students, and the media. The ExaMiner project aims to demonstrate that a system based on this prototype would greatly improve visibility in the research community, the media, prospective students, and government. Looking beyond mapping a single institution, the ExaMiner project also plans to work with SARTRE at Cardiff University to investigate the extra challenge of mapping expertise across Universities.

Project dates

01 Aug 2010

Project funder:

University Research Committee

Project collaborator(s):

Intelligent Systems Laboratory

Department of Computer Science