Author:
You-shan Chung and Keh-Jiann Chen,
Abstract:
The Chinese verb-result compound is productive, but its meaning and syntactic behaviors have posed challenges to theoretical and automatic analyses. Theory-wise, the current study proposes that VRs have inherent affecting direction, which argument mapping principles and selectional restrictions, event structures, or the kinds of semantic/pragmatic principles or real-world knowledge proposed by previous researchers do not seem to account for. Application-wise, we predict the VR’s transitivity and whether the result component is predicated of the logical subject or the logical object, based on the transitivity of individual component verbs and the selectional restrictions between component verbs and arguments. Since the transitivity property and selectional restrictions of individual verbs can be annotated in our lexicon, the rules should fare well in automatic processing. Meanwhile, as the rules have been motivated by linguistic theories and have been observed to make correct predictions in most cases, they are worthy of further large-scale testing.
Keywords:
Verb-result Compound, Transitivity, Lexical-semantics, Meaning Prediction
Author:
Cheng-Ru Li, Chi-Hsin Yu, and Hsin-Hsi Chen
Abstract:
The semantic orientation of terms is fundamental for sentiment analysis in sentence and document levels. Although some Chinese sentiment dictionaries are available, how to predict the orientation of terms automatically is still important. In this paper, we predict the semantic orientation of terms of E-HowNet. We extract many useful features from different sources to represent a Chinese term in E-HowNet, and use a supervised machine learning algorithm to predict its orientation. Our experimental results showed that the proposed approach can achieve 92.33% accuracy.
Keywords:
E-NowNet, Sentiment Analysis, Sentiment dictionary, Semantic orientation, SVM
Author:
Sheng-Fu Wang, Jing-Chen Yang, Yu-Yun Chang, Yu-Wen Liu, and Shu-Kai Hsieh
Abstract:
This study examines how different dimensions of corpus frequency data may affect the outcome of statistical modeling of lexical items. Our analysis mainly focuses on a recently constructed elderly speaker corpus that is used to reveal patterns of aging people’s language use. A conversational corpus contributed by speakers in their 20s serves as complementary material. The target words examined are temporal expressions, which might reveal how the speech produced by the elderly is organized. We conduct divisive hierarchical clustering analyses based on two different dimensions of corporal data, namely raw frequency distribution and collocation-based vectors. When different dimensions of data were used as the input, results showed that the target terms were clustered in different ways. Analyses based on frequency distributions and collocational patterns are distinct from each other. Specifically, statistically-based collocational analysis generally produces more distinct clustering results that differentiate temporal terms more delicately than do the ones based on raw frequency.
Keywords:
Clustering, Collocation, Corpus Linguistics, Temporal Expression, Gerontology
Author:
Bruno Galmar
Abstract:
A morphological family in Chinese is the set of compound words embedding a common morpheme, and Self-organizing maps (SOM) of these Chinese morphological families can be built. Computation of the unified-distance matrices for the SOMs allows us to perform semantic clustering of the members of the morphological families. Such semantic clustering sheds light on the interplay between morphology and semantics in Chinese. We studied how the word lists used in a lexical decision task (LDT) (Chen, Galmar, & Su, 2009) are mapped onto the clusters of the SOMs. We showed that this mapping is helpful to predict whether repetitive processing of members of a morphological family would elicit a satiation in an LDT - habituation - of both morphological and semantic units of the shared morpheme. In their LDT experiment, Chen, Galmar, and Su (2009) found evidence for morphological satiation but not for semantic satiation. Conclusions drawn from our computational experiments and calculations are in accordance with the behavioral experimental results in Chen et al. (2009). Finally, we showed that our work could be helpful to linguists in preparing adequate word lists for behavioral study of Chinese morphological families.
Keywords:
Self-Organizing Maps, Computational Morphology and Semantics