KMi Publications

Tech Reports

Tech Report KMI-06-13 Abstract


A Document-Centric Semantic Annotation Environment to Support Sense-Making
Techreport ID: KMI-06-13
Date: 2006
Author(s): Bertrand Sereno
Download PDF

Prototype Internet infrastructures for scholarly publishing are offering powerful new services over the interconnected ideas and arguments in a literature. However, such services depend on documents being semantically annotated with readers' interpretations, which up until now has been a manual process due to the complexity of such analysis. This thesis investigates the challenge of designing computer-support for document annotation in the context of potentially diverse, contested views about a text's significance, as typifies scholarly research. An interaction design approach is followed to progressively understand the dialogue between the end-users and an appropriate annotation environment. A preliminary analysis of the annotators' goals if followed by an experiment to identify the activities performed in this sense-making task, and a desk research phase, in which approaches to support each of these activities are identified. An active document annotation environment (ClaimSpotter) is then presented. It is built on an open and extensible architecture, which can incorporate new text analysis components as required to overlay annotations onto the original text to draw attention to sections, which may be particularly significant. Facilities to filter and navigate the document in novel ways, to record annotations or reuse existing ones, and to provide pointers to related documents and annotations based on connections mediated by semantic annotations are offered. The tool is finally evaluated in an experimental setting, resulting in a dataset which supported quantitative and qualitative analysis of the end-users' products and process. The analysis characterises how the semantic annotation scheme is used by novices and experts, and how the user interface's rendering of system and end-user annotations shapes interaction. The thesis assesses critically the strengths and weaknesses of the work, providing justification for further cycles of the approach, and concluding with research questions meriting further investigation.

Publication(s):

Sereno, B. (2006). A Document-Centric Semantic Annotation Environment to Support Sense-Making. Unpublished Doctoral Thesis, Knowledge Media Institute, The Open University, Milton Keynes, UK . Submitted May 2005, Approved July 2006 [http://kmi.open.ac.uk]
 
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