The project being reported on is a sub-project of the on-going research of the CKIP (Chinese Knowledge Information Processing) Group. This group was founded by Hsieh Ching-chun in 1986 and is currently directed by Kehjiann Chen and Chu-Ren Huang (Chang et al. 1989, Hsieh et al. 1989, Chen et al. 1991). The CKIP research is divided into three sub-projects according to their goals: 1) An On-line ...

In this paper, we propose an alternative account of linking patterns which does away with intermediary mechanisms such as thematic or actor/undergoer hierarchies. The basis of this theory is constraints on word classes, deened by both syntactic and semantic criteria, which encode generalizations between semantic roles and syntactic arguments. We show that the generalizations a linking theory ne...

In image annotation, the annotation words are expected to represent image content at both visual level and semantic level. However, a single word sometimes is ambiguous in annotation, for example, ”apple” may refer to a fruit or a company. However, when ”apple” combines with ”phone” or ”fruit”, it will be more semantically and visually consistent. In this paper, we attempt to find this kind of ...

Millstream systems have been proposed as a non-hierarchical method for modelling natural language. Millstream configurations represent and connect multiple structural aspects of sentences. We present a method by which the Millstream configurations corresponding to a sentence are constructed. The construction is incremental, that is, it proceeds as the sentence is being read and is complete when...

ONR has been funding Prof. Roy and Prof. Baldwin to perform research on Event Representation in Humans and Machines. Prof. Roy's portion of that research involves developing an unsupervised learning system that acquires structured event representations of human activity grounded in naturalistic video observations. The output of that system is to be an action lexicon that encodes recurrent hiera...

The organizational structure of bilinguals’ linguistic knowledge is a key question in bilingualism research. Research in the last several decades has concentrated on examining whether bilinguals have one shared mental lexicon or two separate lexicons. The evidence tends in favor of the hypothesis that a bilingual’s two languages are distinct at the lexical level, but share a single conceptual l...

The topic of this paper is the development of dynamic lexical representations using artificial neural networks. In previous work on connectionist natural language processing a lot of approaches have experimented with manually encoded lexicon representations for words. However from a cognitive point of view as well as an engineering point of view it is difficult to find appropriate representatio...

The creation of a pronunciation lexicon remains the most inefficient process in developing an Automatic Speech Recognizer (ASR). In this paper, we propose an unsupervised alternative – requiring no language-specific knowledge – to the conventional manual approach for creating pronunciation dictionaries. We present a hierarchical Bayesian model, which jointly discovers the phonetic inventory and...

The original Coffee Taster's Flavor Wheel was developed by the Specialty Coffee Assn. of America over 20 y ago, and needed an innovative revision. This study used a novel application of traditional sensory and statistical methods in order to reorganize the new coffee Sensory Lexicon developed by World Coffee Research and Kansas State Univ. into scientifically valid clusters and levels to prepar...