Articles

Topic models can be used in an unsuper- vised domain adaptation for Word Sense Disambiguation (WSD). In the domain adaptation task, three types of topic mod- els are available: (1) a topic model con- structed from the source domain corpus: (2) a topic model constructed from the tar- get domain …...

Question Classification plays a significant part in Question Answering system. In order to obtain a classifier, we present in this pa- per1 a pragmatic approach that utilizes sim- ple sentence structures observed and learned from the question sentence patterns, trains a set of Finite State Machines (FSM) based on keywords …...

This paper presents a hybrid model for han- dling out-of-vocabulary words in Japanese- to-English statistical machine translation output by exploiting parallel corpus. As the Japanese writing system makes use of four different script sets (kanji, hira- gana, katakana, and romaji), we treat these scripts differently. A machine translitera- tion model …...

In this paper we present a new method for machine learning-based optimization of linguist-written Constraint Grammars. The effect of rule ordering/sorting, grammar- sectioning and systematic rule changes is discussed and quantitatively evaluated. The F-score improvement was 0.41 percentage points for a mature (Danish) tagging grammar, and 1.36 percentage points for …...

Hindi is the lingua-franca of India. Al- though all non-native speakers can com- municate well in Hindi, there are only a few who can read and write in it. In this work, we aim to bridge this gap by building transliteration systems that could transliter- ate Hindi into at-least 7 …...

Wikipedia is an online multilingual encyclopedia that contains a very large number of articles covering most written languages. However, one critical issue for Wikipedia is that the pages in different languages are rarely linked except for the cross-lingual link between pages about the same subject. This could pose serious difficulties …...

Current study is with the aim to identify similarities and distinctions between irony and sarcasm by adopting quantitative sentiment analysis as well as qualitative content analysis. The result of quantitative sentiment analysis shows that sarcastic tweets are used with more positive tweets than ironic tweets. The result of content analysis …...

We propose a method of collective senti- ment classification that assumes dependen- cies among labels of an input set of reviews. The key observation behind our method is that the distribution of polarity labels over reviews written by each user or written on each product is often skewed in the …...

Wu(1994) listed ten words as CFs’ markers in Chinese, they are 早 (early) ， 了 (perfect/perfective marker), 要不是/要不然 (had it not been the case), 没(didn’t), 就好了 (would have been great if only)，还以为(had thought), 原来应该(should have been), ...的 话 (in the case), 真 的 (really). However, according to our definitions, none …...

GLÀFF, a Large Versatile French Lexicon This paper introduces GLÀFF, a large-scale versatile French lexicon extracted from Wiktionary, the collaborative online dictionary. GLÀFF contains, for each entry, a morphosyntactic description and a phonetic transcription. It distinguishes itself from the other available lexicons mainly by its size, its potential for constant …...

This paper proposes two syntax-semantics correspondence rules which consistently ac- count for the distribution of Japanese loca- tive postpositions ni and de. We demon- strate how to adapt the machinery of the oc- currence of the postpositions based on the assumption of Conceptual Semantics (Jack- endoff, 1983; 1990; 1991) to …...

This study investigates the conceptualization of our bodily orientation in a quantitative corpus-based approach of collostructional analysis. Based on the symbolic nature of constructions, we examine the correlation patterns of the covarying collexeme NPs and 13 major spatial particles in English Preposition Construction through exploratory statistical methods. The distributional patterns …...

This paper aims to introduce the Korean Sentiment Analysis Corpus named KOSAC. KOSAC is a corpus consisting of 332 news articles taken from the Sejong Syntactic Parsed Corpus. These sentences have been manually-tagged for sentimental features. The corpus includes 7,713 sentence subjectivity tags and 17,615 opinionated expression tags based on …...

This study investigates of-constructions in the predicates of two verbs, demonstrate and show, in academic discourse. A construction perspective is taken to examine how the two predicate constructions (‘demonstrate N1 of N2’ and ‘show N1 of N2’) would differ when the information-weighting of N1 and N2 are considered. The noun …...

This study examines the sound profiles of sites of initiation in French and Mandarin recycling repair (also disfluent repetitions). 150 examples of disfluent repetitions were extracted from comparative speech corpora of naturally occurred, face-to-face Mandarin and French interaction. By the approach of interactional prosody plus impressionistic judgments, each instance of …...

Newspapers remain an important media from which people obtain a wide variety of information. In Japan, there are five major newspapers, having their own opinions and ideologies. Although these are readily recognized, they are infrequently investigated from the viewpoint of their textual characteristics. This study analyzes these differences among the …...

It is important for a natural language dia- logue system to interpret relations among event concepts appearing in a dialogue. The more complex a dialog becomes, the more essential it becomes for a natural language dialogue system to perform this kind of in- terpretation. Traditionally, many studies have focused on …...

Functional Data Analysis (FDA) is used to investigate Tone 3 sandhi in Taiwan Mandarin. Tone 3 sandhi is a tone change phenomenon that arises when two low tones occur in succession resulting in the first tone being realised as a rising tone. Tone dyads T2T3 and T3T3 were compared in …...

This paper investigates the formal seman- tics of reduplication in Cantonese, i.e. how the meaning of reduplicated forms are en- coded and computed with the given mean- ing from the base forms. In particular, this paper argues that reduplication denotes a summation function that adds up argu- ments (be they …...

This paper presents work on the semantic annotation of a multimodal corpus of English television news. The annotation is performed on the second-by-second- aligned transcript layer, adding verb frame categories and semantic roles on top of a morphosyntactic analysis with full dependency information. We use a rule- based method, where …...