Processing of cyclic graphs with recursive neural networks
MetadataShow full item record
CitationRezaei, Z. (2009). Processing of cyclic graphs with recursive neural networks. Maltepe Üniversitesi. s. 402.
Recursive neural networks are a powerful tool for processing structured data. They are filling the gap between connectionism, which is usually related to poorly organized data, and a great variety of real-world problems such as document processing, where the information is naturally incoded in the relationships among the basic entities. More precisely recursive neural networks can deal only with directed ordered acyclic graphs (DOAGs), in which the children of any given node are ordered. While this assumption is reasonable in some applications, it introduces unnecessary constraints in others. Example of such applications is classification of HTML pages. In this paper, we explain a methodology, which allow us to process any cyclic directed graph. The computational power of recursive neural networks is established.
SourceInternational Conference of Mathematical Sciences
- Makale Koleksiyonu 
The following license files are associated with this item: