Cumincad : CUMINCAD Papers : Search Results

CumInCAD is a Cumulative Index about publications in Computer Aided Architectural Design
supported by the sibling associations ACADIA, CAADRIA, eCAADe, SIGraDi, ASCAAD and CAAD futures

PDF papers
References

Hits 1 to 2 of 2

Reformat results as: short short into frame detailed detailed into frame

_id	bd13
authors	Martens, B., Turk, Z. and Cerovsek, T.
year	2001
title	Digital Proceedings: Experiences regarding Creating and Using
doi	https://doi.org/10.52842/conf.ecaade.2001.025
source	Architectural Information Management [19th eCAADe Conference Proceedings / ISBN 0-9523687-8-1] Helsinki (Finland) 29-31 August 2001, pp. 25-29
summary	This paper describes the developments of the CUMINCAD database since 1999 when it was first presented and some statistical information, how the service is being used. CUMINCAD started as a bibliographic database storing meta information about CAADrelated publications. Recently, full texts are being added. The process of creation of electronic copies of papers in pdf-format is described as well as decisions which were taken in this context. Over the last two years 20.000 users visited CUMINCAD. We present a brief analysis of their behavior and interaction patterns. This and the forthcoming possibility of a full-text-search will open up a new perspective for CAAD-research.
keywords	CAAD-Related Publications, Web-Based Bibliographic Database, Searchable Index, Retrospective CAAD Research, Purpose Analysis
series	eCAADe
email
last changed	2022/06/07 07:59

_id	8d50
authors	Turk, Z., Cerovsek, T. and Martens, B.
year	2001
title	The Topics of CAAD. A Machine's Perspective
source	Proceedings of the Ninth International Conference on Computer Aided Architectural Design Futures [ISBN 0-7923-7023-6] Eindhoven, 8-11 July 2001, pp. 547-560
summary	Ontology of a scientific field typically includes a taxonomy that breaks up the field into several topics. The break-up is present in the organisation of information in books, libraries and on the Web. An on-line database of papers related to CAAD called CUMINCAD was created and it includes over 3000 papers with abstracts. They are available through the search interface - one knows an author or a keyword and can find the papers where such keyword or author's name appears. Alternative interface would be through browsing papers topic by topic. The papers, however, are not categorised. In this paper, we present the efforts to use the machine learning and data mining techniques to automatically group the papers into clusters and create a set of keywords that would label a cluster. The hypothesis was that an algorithm would create clusters of papers automatically and that the clusters would be similar to the groupings a human would have made. We investigated several algorithms for doing an analysis like that but were unable to prove the original hypothesis. We conclude that it requires more than objective statistical analysis of the words in abstracts to create an ontology of CAAD.
keywords	CAAD, Machine Learning, Clustering, Pattern Recognition
series	CAAD Futures
email
last changed	2006/11/07 07:22

No more hits.