KMi Seminars
Information-Theoretic Semantic Multimedia Indexing
This event took place on Wednesday 27 June 2007 at 11:30

 
Joćo Magalhćes Imperial College London, and KMi, The Open University

To solve the problem of indexing collections with diverse text documents, image documents, or documents with both text and images, one needs to develop a model that supports heterogeneous types of documents.
In this paper, we show how information theory supplies us with the tools necessary to develop a unique model for text, image, and text/image retrieval. In our approach, for each possible query keyword we estimate a maximum entropy model based on exclusively continuous features that were pre-processed. The unique continuous feature-space of text and visual data is constructed by using a minimum description length criterion to find the optimal feature-space representation (optimal from an information theory point of view). We evaluate our approach in three
experiments: only text retrieval, only image retrieval, and text combined with image retrieval.

 
KMi Seminars
 

Knowledge Management is...


Knowledge Management
Creating learning organisations hinges on managing knowledge at many levels. Knowledge can be provided by individuals or it can be created as a collective effort of a group working together towards a common goal, it can be situated as "war stories" or it can be generalised as guidelines, it can be described informally as comments in a natural language, pictures and technical drawings or it can be formalised as mathematical formulae and rules, it can be expressed explicitly or it can be tacit, embedded in the work product. The recipient of knowledge - the learner - can be an individual or a work group, professionals, university students, schoolchildren or informal communities of interest.
Our aim is to capture, analyse and organise knowledge, regardless of its origin and form and make it available to the learner when needed presented with the necessary context and in a form supporting the learning processes.