Showing papers by "Chris Mellish published in 1997"

Dynamic Generation of Museum Web Pages: The Intelligent Labelling Explorer

Abstract: The first phase of the Intelligent Labelling Explorer project has built the ILEX-1.1 system, which uses artificial intelligence technology to generate descriptions of objects displayed in a museum gallery. Each description appears on a World Wide Web page, and the user can move from page to page, viewing the objects in any order, mimicking the experience of someone walking through the museum. Crucially, these descriptions aren't simply retrieved from a storage space, but are generated on demand by combining canned text with fully generated text in a coherent way. This use of dynamic hypertext allows ILEX-1.1 to generate descriptions appropriate to the expertise level of the user and to refer back to objects the user has already seen or to suggest objects the user might be interested in based on what objects they've chosen to look at so far. This paper discusses the advantages of dynamic hypertext and issues related to generating a text that hangs together well.

An empirical study on the generation of anaphora in Chinese

Abstract: The goal of this work is to study how to generate various kinds of anaphora in Chinese, including zero, pronominal, and nominal anaphora, from the syntactic and semantic representation of multisentential text. In this research we confine ourselves to descriptive texts. We examine the occurrence of anaphora in human-generated text and those generated by a hypothetical computer equipped with anaphor generation rules, assuming that the computer can generate the same texts as the human except that anaphora are generated by the rules. A sequence of rules using independently motivated linguistic constraints is developed until the results obtained are close to those in the real texts. The best rule obtained for the choice of anaphor type makes use of the following conditions: locality between anaphor and antecedent, syntactic constraints on zero anaphora, discourse segment structures, salience of objects and animacy of objects. We further establish a rule for choosing descriptions if a nominal anaphor is decided on. We have implemented the above rules in a Chinese natural language generation system that is able to generate descriptive texts. We sent some generated texts to a number of native speakers of Chinese and compared human-created results and computer-generated text to investigate the quality of the generated anaphora. The results of the comparison show that the rules are fairly effective in dealing with the generation of anaphora in Chinese.

Exploring a gallery with intelligent labels

Abstract: Ordinary labels on museum or gallery objects compromise the competing demands of diiering visitors, curators and educators. Although adaptive hypermedia systems ooer a degree of personalisation to the visitor, they do not go as far as they might. Intelligent labelling can achieve higher levels of personalisation, and this is desirable because it leads to a more coherent and educational visit, and because the overall interaction with the hypermedia resource becomes more like an active conversation, and less like reading a static book. We use natural language generation techniques, from the eld of artiicial intelligence, to generate personalised labels on demand. Our system can deliver these labels as Web pages, in an electronic gallery, or as synthesised speech in a physical gallery. We discuss aspects of coherence and conversa-tionality, and illustrate them with a simple case.