External Publications

2012 [88]

2011 [136]
2010 [191]

2009 [166]

2008 [191]

2007 [145]

Earlier Publications [788]

Tech Reports

2012 [3]

2011 [4]

2010 [3]

2009 [6]

2008 [8]

2007 [4]

Earlier Tech Reports [197]

Books [8]

JIME

KMi Tools



 

News

 
KMi Publications

Tech Reports

Tech Report kmi-97-22 Abstract


Parameter Estimation in Bayesian Networks from Incomplete Databases
Techreport ID: kmi-97-22
Date: 1997
Author(s): Marco Ramoni and Paola Sebastiani
Download Postscript

Current methods to learn Bayesian Networks from incomplete databases share the common assumption that the unreported data are missing at random. This paper describes a method - called Bound and Collapse (BC) - to learn Bayesian Networks from incomplete databases which allows the analyst to efficiently integrate the information provided by the database and the exogenous knowledge about the pattern of missing data. BC starts by bounding he set of estimates consistent with the available information and then collapses the resulting set to a point estimate via a convex combination of the extreme points, with weights depending on the assumed pattern of missing data. Experiments comparing BC to the Gibbs Samplings are also provided. 1. Knowledge Media Institute, The Open University. 2. Department of Actuarial Science and Statistics, City University.
 
KMi Publications
 

Multimedia and Information Systems is...


Multimedia and Information Systems
Our research is centred around the theme of Multimedia Information Retrieval, ie, Video Search Engines, Image Databases, Spoken Document Retrieval, Music Retrieval, Query Languages and Query Mediation.

We focus on content-based information retrieval over a wide range of data spanning form unstructured text and unlabelled images over spoken documents and music to videos. This encompasses the modelling of human perception of relevance and similarity, the learning from user actions and the up-to-date presentation of information. Currently we are building a research version of an integrated multimedia information retrieval system MIR to be used as a research prototype. We aim for a system that understands the user's information need and successfully links it to the appropriate information sources, be it a report or a TV news clip. This work is guided by the vision that an automated knowledge extraction system ultimately empowers people making efficient use of information sources without the burden of filing data into specialised databases.

Visit the MMIS website