Caroline Barrire, From a Children's First Dictionary to a Lexical Knowledge Base of Conceptual Graphs, Ph.D thesis, School of Computing Science, Simon Fraser University, 1997. Available at ftp://www.cs.sfu.ca/pub/cs/nl/BarrierePhD.ps.gz

 Home/Search   Document Details and Download   Summary   Related Articles  

This paper is cited in the following contexts:
Learning to Generate CGs from Domain Specific Sentences - Zhang, Yu   (Correct)

....of link grammar and the semantic knowledge of domain is binded by training. This coupling is loose and can be changed by training in different domains. Another kind of technique advanced in previous work is to directly map between syntatic structure and semantic structure of CG such as [15] and [16]. We call them structure mapping. In this respect, they are more similar to our work. To map to more flat strucutres of conceptual graphs, 15] uses syntactic predicates to represent the grammatical relations in the parse tree. Instead, in our work, link grammar is employed to directly obtain a ....

....to a database of translating rules. All the possibilities are finally checked against a LKB. Different from [15] s approach, our work doesn t uses manual rules. Moreover, we separate the semantic mapping into several steps which greatly reduce the total number of possibilities. In another work in [16], parse tree is first mapped to a syntactic CG . The syntactic CG is then mapped to a real CG. This approach again heavily uses manually constructed mapping rules (such as Parse To CG rules and SRTG rules in [16] Further more, unlike [16] s two tier mapping, we do the mapping from syntactic ....

[Article contains additional citation context not shown here]

Caroline Barrire, From a Children's First Dictionary to a Lexical Knowledge Base of Conceptual Graphs, Ph.D thesis, School of Computing Science, Simon Fraser University, 1997. Available at ftp://www.cs.sfu.ca/pub/cs/nl/BarrierePhD.ps.gz

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC