Mintram, R. and Vincent, J., 2003. A General Framework for the Transformation of Structured Data into Vector Representations. In: 21st IASTED International Conference on Applied Informatics (AI 2003), 10 – 13 February 2003, Innsbruck, Austria, pp. 79-84.
Full text not available from this repository.
A consequence of the connectionist approach to artificial intelligence is the requirement for structured data to be encoded into fixed width vector representations. This paper presents a generalised approach to this problem that views the process of encoding and decoding vector representations as separate mechanisms.
|Item Type:||Conference or Workshop Item (Paper)|
|Additional Information:||There are few paradigms for extracting features from structured data. This research distills the common principles from these procedures and shows that they are a special case of a general approach to structured data representation. This is achieved by proposing generic algorithms which can be instantiated into extisting extraction metyhods. In particular, the RAAM and SRAAM neural network techniques are shown to be generically identical. Furthermore, it is argued that any as yet unspecied techniques will also confom to this generic paradigm.|
|Group:||School of Design, Engineering & Computing > Software Systems Research Centre|
|Deposited By:||INVALID USER|
|Deposited On:||20 Feb 2007|
|Last Modified:||07 Mar 2013 14:35|
|Repository Staff Only -|
|BU Staff Only -|
|Help Guide -||Editing Your Items in BURO|