In this paper , <term> events </term> are defined as <term> event terms </term> and <term> associated event elements </term> .
Finally , we show how this new <term> tagger </term> achieves state-of-the-art results in a <term> supervised , non-training intensive framework </term> .
We have built and will demonstrate an application of this approach called <term> LCS-Marine </term> .
The lack of automatic <term> methods </term> for <term> scoring system output </term> is an impediment to progress in the field , which we address with this work .
In order to judge three types of the <term> errors </term> , which are characters wrongly substituted , deleted or inserted in a <term> Japanese bunsetsu </term> and an <term> English word </term> , and to correct these <term> errors </term> , this paper proposes new methods using <term> m-th order Markov chain model </term> for <term> Japanese kanji-kana characters </term> and <term> English alphabets </term> , assuming that <term> Markov probability </term> of a correct chain of <term> syllables </term> or <term> kanji-kana characters </term> is greater than that of <term> erroneous chains </term> .
To deal with this <term> complexity </term> , we describe how <term> disjunctive </term> values can be specified in a way which delays <term> expansion </term> to <term> disjunctive normal form </term> . This paper describes a domain independent strategy for the <term> multimedia articulation of answers </term> elicited by a <term> natural language interface </term> to <term> database query applications </term> .
In this paper , we present <term> SPoT </term> , a <term> sentence planner </term> , and a new methodology for automatically training <term> SPoT </term> on the basis of <term> feedback </term> provided by <term> human judges </term> .
In our experiment , the method could construct a <term> corpus </term> consisting of 126,610 <term> sentences </term> . This paper examines what kind of <term> similarity </term> between <term> words </term> can be represented by what kind of <term> word vectors </term> in the <term> vector space model </term> .
Manual acquisition of <term> semantic constraints </term> in broad domains is very expensive . This paper presents an <term> automatic scheme </term> for collecting <term> statistics </term> on <term> co-occurrence patterns </term> in a large <term> corpus </term> .
<term> Combination methods </term> are an effective way of improving <term> system performance </term> . This paper examines the benefits of <term> system combination </term> for <term> unsupervised WSD </term> .
In this paper , we report a system <term> FROFF </term> which can make a fair copy of not only texts but also graphs and tables indispensable to our papers .
By holding multiple <term> candidates </term> for <term> understanding </term> results and resolving the <term> ambiguity </term> as the <term> dialogue </term> progresses , the <term> discourse understanding accuracy </term> can be improved . This paper proposes a method for resolving this <term> ambiguity </term> based on <term> statistical information </term> obtained from <term> dialogue corpora </term> .
The characteristics of this method is that it is fully automatic and can be applied to arbitrary <term> HTML documents </term> .
In this presentation , we describe the features of and <term> requirements </term> for a genuinely useful <term> software infrastructure </term> for this purpose .
Participants should be able , after attending this workshop , to set out building an <term> SMT system </term> themselves and achieving good <term> baseline results </term> in a short time .
In this paper , we present a <term> corpus-based supervised word sense disambiguation ( WSD ) system </term> for <term> Dutch </term> which combines <term> statistical classification ( maximum entropy ) </term> with <term> linguistic information </term> .
In this paper , we present our work on the detection of <term> question-answer pairs </term> in an <term> email conversation </term> for the task of <term> email summarization </term> .
The <term> board </term> plugs directly into the <term> VME bus </term> of the <term> SUN4 </term> , which controls the system and contains the <term> natural language system </term> and <term> application back end </term> . This paper reports on two contributions to <term> large vocabulary continuous speech recognition </term> .
First , how <term> linguistic concepts </term> are acquired from <term> training examples </term> and organized in a <term> hierarchy </term> ; this task was discussed in previous papers [ Zernik87 ] .
Interpreting <term> metaphors </term> is an integral and inescapable process in <term> human understanding of natural language </term> . This paper discusses a <term> method of analyzing metaphors </term> based on the existence of a small number of <term> generalized metaphor mappings </term> .
hide detail