CIPHER Project Summary - The Technical Goals

Collaborative discovery tools to support members of the community in the creation of personal and communal archives

The Community Story Exchange was implemented and tested in two applications (Bletchley Park guides and Scooter, described in section 3.4). Similar support for the creation of personal and communal archives is provided in the CH-Portal, CH-Publisher and CH-Editor suite of tools from CTU, and the Explorer Content Management Tools from DIT. Like the Community Story Exchange these tools were tested in the two trials of their respective forums.

Navigation and exploration facilities based on self-organizing maps

This technology was fully developed by UIAH. UIAH applied the SOM techniques both to their own content and to the content supplied by other partners. Based on the UIAH ideas and experiences, KMI have developed and tested a small experimental SOM tool.

Figure 7. A SOM similarity cluster map of the Bestiarium Construendum ontology

Ontology driven dynamic narratives, providing coherent, compelling stories according to visitor interests and needs

Different versions of this technology were developed and tested by a number of partners. Story Fountain was developed at the Open University and used both by the Open University and the Czech Technical University. DIT have integrated these ideas into their Irish Explorer tool, UIAH approached the problem from the viewpoint of soft ontologies. As a basis we have reused the existing ISO standard proposal called CRM ontology which was robust enough and saved us a lot of time. This ontology is the still being used in CIPHER applications and we are in contact with their authors and provide feedback from our experiences.

Developer tools to support the construction of CH Forums.

An extensive set of tools have been developed and shared among the project partners. They include tools for developing ontologies, annotating digital heritage resources including pictures (see Figure 3 as an example), organising content and generating narratives and publishing. These tools can be divided into two groups: stand-alone tools and integrated tools supporting needs of specific community. Stand-alone tools are documented as appendices to deliverable D26 and can be easily used by third parties.

Maintenance tools to support CH Forum evolution, including visitor use analysis.

Some of the tools developed allow the user to download and maintain existing content. Since the second trials required the integration of a large quantity of new content this feature has been extensively tested. Feedback from the analysis of the users' behaviour is used to monitor the performance of the tools. For example, the visitors' use of the Bletchley Park mobile information service is logged and made available to the Bletchley Park tour guides (an example screen view is shown in Figure 8).

Figure 8. Evaluating visitors' activity in Bletchley Park

Interface templates employing technologies to combine, position and present media elements in contextually appropriate ways to a range of hardware clients, including mobile devices

This goal has been addressed in a different ways according to the needs of each application. For example, in the Community Story Exchange tool a number of templates have been developed to support different models of communication. For example, a "fan model" with a single problem setting, many independent and for limited time hidden answers and final plenary has been implemented, see Figure 9. This model is appropriated for supporting communication which needs independent action on a single question, e.g. a puzzle. Other supported models include "stand-alone" where the contributions are independent and "chain where the contributions follow a sequential ordering.

Figure 9. Communication and presentation model "fan".

The presentation templates allow integration of a wide spectrum of different media and digital resources. For example, the Irish Explorer 3D presentation tool makes it possible to integrate web accessible 3D presentations and linking them to the Explorer knowledge base. An example of an interactive 3D presentation of Fourknocks passage tomb in Co. Meath, Ireland is shown in Figure 10.

Figure 10. An example of an integrated 3D presentation available in the Irish Forum

Mobile interactive technologies were developed and integrated with the Bletchley Park Forum exploration tools (as described in section 3.4). The idea is illustrated in Figure 11. During or after the visit the visitor sends an SMS message to a mobile telephone number associated with the server. The message contains keywords relating to the content the visitor would like to explore in depth. When returning home, the visitor connects to the web server and provides the phone number from which the message was sent. All of the relevant information is then made available and organised in a format to facilitate further exploration.

Figure 11. Mobile technology at Bletchley Park

Language technologies to support multilingual access to resources, through support for human translation and context sensitive cross language associations

Current language technologies do not allow us to automatically translate text with sufficient quality of translation, especially in such a language sensitive domain as cultural heritage. In many cases, language itself is a part of the heritage content. However, in order to facilitate the access to heritage objects a simplified form of multilingual support has been tested using conceptual graphs. The idea is that the user will acquire the basic grasp of the content and if interested, more expensive professional translation can be requested by the community.

Summaries of a number of documents have been represented as conceptual graphs and translated into several languages. A group of people were presented with the conceptual graph and asked to write down the summary in English. The findings indicated that a sufficient basic understanding was achieved, with only about 15% of the participants extrapolating the content and describing events that were not included in the original document (e.g. "there was a fire in a castle" was interpreted as "the castle was destroyed by fire"), and about 15% of respondents omitting one or more of the summarised facts. This technology is being studied further by the Czech Technical University.