We would like to build a model of semantic knowledge that have the capacity to acquire and represent semantic information that is ambiguous, vague and incomplete. Furthermore, the model should be able to acquire this knowledge in an unsupervised fashion from unstructured text data. Such a model needs to be both highly adaptive and very robust. In this submission, we will first try to identify s...

In his logic of conversation, Paul Grice made an important distinction between two kinds of act that speakers may perform when uttering a sentence and, correspondingly, two kinds or levels of speaker meaning: (i) saying, hence what is said, and (ii) implicating, hence what is implicated (implicature) (Grice 1975). The distinction between what is said and what is implicated can be seen as one in...

A distributed representation has become a popular approach to capturing a word meaning. Besides its success and practical value, however, questions arise about the relationships between a true word meaning and its distributed representation. In this paper, we examine such a relationship via polymodal embedding approach inspired by the theory that humans tend to use diverse sources in developing...

As a testbed of statistical word aligner, we implemented the prototype of statistical word aligner by graphical models [2, 10]. The advantage of using graphical method resides in its extensibility compared to the traditional approach for statistical word alignment [3, 22, 14]. Although there are semi-supervised word aligner [6], we only talk about unsupervised word aligner [3, 22, 14]. The capa...

Two eye-tracking experiments investigated the processing of mass nouns used as count nouns and count nouns used as mass nouns. Following Copestake and Briscoe (1995), the basic or underived sense of a word was treated as the input to a derivational rule (“grinding” or “portioning”) which produced the derived sense as output. It was hypothesized that in the absence of biasing evidence readers wo...

AUTOMATIC ACQUISITION OF WORD MEANING FROM CONTEXT by Peter Mark Hastings Chair: Assistant Professor Steven Lytinen This thesis presents an automatic, incremental lexical acquisition mechanism that uses the context of example sentences to guide inference of the meanings of unknown words. The goal of this line of research is to allow a Natural Language Processing (NLP) system to cope with words ...

Contrasting meaning is a basic aspect of semantics. Recent word-embedding models based on distributional semantics hypothesis are known to be weak for modeling lexical contrast. We present in this paper the embedding models that achieve an F-score of 92% on the widely-used, publicly available dataset, the GRE “most contrasting word” questions (Mohammad et al., 2008). This is the highest perform...

Models of word meaning, built from a corpus of text, have demonstrated success in emulating human performance on a number of cognitive tasks. Many of these models use geometric representations of words to store semantic associations between words. Often word order information is not captured in these models. The lack of structural information used by these models has been raised as a weakness w...

Strongly reduced word forms are difficult to interpret without syntactic or semantic context. When speakers of Dutch are presented with the word mok in isolation, they tend not to be able to assign a meaning to this string of phonemes. It is only when the word is embedded in a sentence that its meaning becomes available. Interestingly, listeners who understood the meaning of MOK tend to think t...

provided the first quantitative comparison of related languages based on a well-defined model of language change. The stochastic nature of this model was poorly understood by linguists, in the main, and many have rejected the theory in the course of a protracted and confused controversy. Meanwhile field linguists, especially those working with language groups of unknown history, have accepted l...