Marie-Claude Landau. Solving Ambiguities in the Semantic Representation of Texts. In Tenth International Workshop on Expert Systems and Their Applications, pages 119--130, 1990.  Home/Search   Document Details and Download   Summary   Related Articles

....per sentence, we can ask the user to manually choose one graph. Many ideas presented in this chapter are not novel. But the exercise of performing a sentence to conceptual graph transformation allows the presentation of a good synthesis of multiple NLP issues. In some papers using CGs for NLP [61, 88], the authors assume the sentence to graph transformation is already done, but they never really explain how it is done, with what success and what effect how well the transformation is done has on subsequent results. The reader will see that a sentence to CG transformation is not an easy task. It ....

....we will look into the graph definitions of eat and fork to find the subgraph [eat] with) fork] We can choose the best structural interpretation depending on the similarity between our graph definitions. Finding similarity with graph definitions can also help in the case of anaphora resolution [88]. One interesting research based on an LKB of Conceptual Graphs is the project KALIPSOS described in [61] which performed some natural language understanding tasks using CGs. They differentiate between two aspects of a word, its canonical graph and its meaning by use. Their LKB consists of a list ....

Marie-Claude Landau. Solving Ambiguities in the Semantic Representation of Texts. In Tenth International Workshop on Expert Systems and Their Applications, pages 119--130, 1990.

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