Search

In spoken dialogue systems, a robust treatment is required for the input utterance due to the recognition errors and a large set of spontaneous speech phenomena. This paper describes a framework for robust analysis and interpretation of spoken language. This framework includes inherently robust mechanisms, i.e., robust semantic parsing and robust interpretation, and an additional component as well, i.e., recognition error correction. A preliminary evaluation demonstrates that a good robustness is achieved efficiently under this framework. [ABSTRACT FROM AUTHOR]

The natural language interpretation in Intensional Logical Model (ILM) consists of three steps: basic meaning description, concepts combination and situational description. This paper focuses on the general approach to and process of situational description, including the application of the ILM approach to the conceptual combination in the field of dialogue. In the paper, the mobile phone operation is an instantiation of the human-computer interaction. In the first section, the task for the situational description will be outlined. In the second section, the dynamic and static conceptual combination will be discussed in the framework of ILM. And the computing model will be introduced for situational description in third section. [ABSTRACT FROM AUTHOR]

This article proposes a neural network for building Chinese and English semantic resources connection. Abundant monolingual semantic information is stored into its bipartite graph structure respectively. Two hidden layers are also set in every part, word layer and concept layer. Every word associates with different concepts separately; every concept includes different vocabularies; and these two layers also independently connect to their counterparts through bipartite graph. These distributed characteristics in hidden layers meet the need of parallel network computing. The unsupervised method is used to train the network, and samples are translation lexicons, results of the bilingual word-level alignment algorithm. The training principle comes from the inspiration of bilingual semantic asymmetry. Every translational equivalent contains the unambiguous information by comparison between source and target languages. These translation lexicons are viewed as a kind of special context. They almost have definite meaning. Every input will activate and suppress various kinds of potential connections by the interaction of hidden layers, and modify their connective weights. Finally a demo test presents. [ABSTRACT FROM AUTHOR]

In this paper, a machine readable dictionary is utilized to acquire Chinese noun pairs satisfying five lexical relations. For low accuracy of current Chinese parser, our method is different from the traditional ones that use parsing firstly. The new method is designed to be a three-step procedure. Firstly, it annotates the paraphrase of some nominal entries that are used as training data. Secondly, patterns that denote lexical relations between nouns are defined and the applicability of the patterns is learnt from training Maximum Entropy model. At last, these patterns are applied to the remaining portion of the dictionary. A relatively satisfying result is achieved. [ABSTRACT FROM AUTHOR]