Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/iiaiaai55812.2022.00097

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper (scientific journal)  

J-GLOBAL

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Authorship：Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)  

J-GLOBAL

researchmap

Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Association for Computational Linguistics  

researchmap

Other Link： https://dblp.uni-trier.de/conf/paclic/2018

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

4 Rounds (4R) training method is practiced in industrial office site for reducing accidents caused by human factors. The 4R method enables to raise hazard-prediction capability of worker such as coping, decision-making to avoid danger situation. The workers as trainees train on their own by finding hazards which lurked in the hazard prediction training (KYT in Japanese) sheet. However, there is a large problem that a single trainee cannot train oneself using 4R method because the training of 4R method needs instruction of expert as human instructor.
To solve that problem, we aim to develop hazard prediction training system. The advantage of this system enables trainee to train oneself anytime/anywhere using 4R method. In this research paper, we reports about our proposal of training system, the development of subsystem which based on machine learning to evaluate trainee's answer correct or not, and reports the result of evaluation experimental that showed the average accuracy was 63.0 +/- 22.9 [%].

DOI： 10.1109/IIAI-AAI.2015.301

Web of Science

researchmap

Publishing type：Research paper (scientific journal)   Publisher：International Institute of Applied Informatics  

DOI： 10.52731/iee.v1.i3.49

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：INSTICC-INST SYST TECHNOLOGIES INFORMATION CONTROL & COMMUNICATION  

This paper presents a system for word sense disambiguation based on a manually constructed thesaurus of predicate-argument structure, which is an ontology on the linguistic side providing essential information for mapping form texts to verb concepts. This system can be effective for word sense disambiguation even though the target word sense system is different from the thesaurus. We applied the proposed word sense disambiguation system to the test corpus of SemEval-2010 Japanese tasks. Experimental results showed that the thesaurus-based disambiguation system outperformed a CRFs-based system in recall rates of verb sense disambiguation. From the results of verb sense disambiguation, we clarified that the abstracted verb classes (709 types) in our proposed system were effective sets for verb sense disambiguation.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

Current research in text mining favors the quantity of texts over their representativeness. But for bilingual terminologymining, and for many language pairs, large comparable corpora are not available. More importantly, as terms are defined vis-à-vis a specific domain with a restricted register, it is expected that the representativeness rather than the quantity of the corpus matters more in terminology mining. Our hypothesis, therefore, is that the representativeness of the corpus is more important than the quantity and ensures the quality of the acquired terminological resources. This article tests this hypothesis on a French-Japanese bilingual term extraction task. To demonstrate how important the type of discourse is as a characteristic of the comparable corpora, we used a state-of-the-art multilingual terminology mining chain composed of two extraction programs, one in each language, and an alignment program. We evaluated the candidate translations using a reference list, and found that taking discourse type into account resulted in candidate translations of a better quality even when the corpus size was reduced by half. © 2010 ACM.

DOI： 10.1145/1839478.1839479

Scopus

researchmap

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

The extraction of verb synonyms is a key technology to build a verb dictionary as a language resource. This paper presents a co-clustering-based verb synonym extraction approach that increases the number of extracted meanings of polysemous verbs from a large text corpus. For verb synonym extraction with a Clustering approach dealing with polysemous verbs can be one problem issue because each polysemous verb should be categorized into different clusters depending on each meanings thus there is a high possibility of failing to extract sonic of the meanings of polysemous verbs. Our proposed approach can extract the different meanings of polysemous verbs by recursively eliminating the extracted Clusters from the initial data set. The experimental results of verb synonym extraction show that the proposed approach increases the correct verb clusters by about 50% with a 0.9% increase in precision and a 1.5% increase in recall over the previous approach.

DOI： 10.1587/transinf.E92.D.2334

Web of Science

researchmap

Publishing type：Research paper (scientific journal)  

researchmap

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：OXFORD UNIV PRESS  

BioCaster is an ontology-based text mining system for detecting and tracking the distribution of infectious disease outbreaks from linguistic signals on the Web. The system continuously analyzes documents reported from over 1700 RSS feeds, classifies them for topical relevance and plots them onto a Google map using geocoded information. The background knowledge for bridging the gap between Layman's terms and formal-coding systems is contained in the freely available BioCaster ontology which includes information in eight languages focused on the epidemiological role of pathogens as well as geographical locations with their latitudes/longitudes. The system consists of four main stages: topic classification, named entity recognition (NER), disease/location detection and event recognition. Higher order event analysis is used to detect more precisely specified warning signals that can then be notified to registered users via email alerts. Evaluation of the system for topic recognition and entity identification is conducted on a gold standard corpus of annotated news articles.

DOI： 10.1093/bioinformatics/btn534

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE COMPUTER SOC  

This paper describes that a graph-based co-clustering approach is suitable for extraction of verb synonyms from large scale texts. The proposed bipartite graph algorithm can produce clusters of verb synonyms as well as noun synonyms taking into account word co-occurrence between verb and its argument. Experimental results show that the co-clustering approach achieve higher accuracy than those by a vector-based single clustering approach that are usually used for construction of thesaurus.

DOI： 10.1109/ISUC.2008.66

Web of Science

researchmap

researchmap

researchmap

Publisher：The Association for Computational Linguistics  

CiNii Article

researchmap

Other Link： http://dblp.uni-trier.de/db/conf/acl/acl2007.html#conf/acl/MorinDTK07

Language：English   Publishing type：Research paper (scientific journal)   Publisher：SPRINGER  

A lack of surveillance system infrastructure in the Asia-Pacific region is seen as hindering the global control of rapidly spreading infectious diseases such as the recent avian H5N1 epidemic. As part of improving surveillance in the region, the BioCaster project aims to develop a system based on text mining for automatically monitoring Internet news and other online sources in several regional languages. At the heart of the system is an application ontology which serves the dual purpose of enabling advanced searches on the mined facts and of allowing the system to make intelligent inferences for assessing the priority of events. However, it became clear early on in the project that existing classification schemes did not have the necessary language coverage or semantic specificity for our needs. In this article we present an overview of our needs and explore in detail the rationale and methods for developing a new conceptual structure and multilingual terminological resource that focusses on priority pathogens and the diseases they cause. The ontology is made freely available as an online database and downloadable OWL file.

DOI： 10.1007/s10579-007-9019-7

Web of Science

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Elsevier  

Objective: Support vector machines (SVMs) have achieved state-of-the-art performance in several classification tasks. In this article we apply them to the identification and semantic annotation of scientific and technical terminology in the domain of molecular biology. This illustrates the extensibility of the traditional named entity task to special domains with large-scale terminologies such as those in medicine and related disciplines. Methods and materials: The foundation for the model is a sample of text annotated by a domain expert according to an ontology of concepts, properties and relations. The model then learns to annotate unseen terms in new texts and contexts. The results can be used for a variety of intelligent language processing applications. We illustrate SVMs capabilities using a sample of 100 journal abstracts texts taken from the {human, blood cell, transcription factor} domain of MEDLINE. Results: Approximately 3400 terms are annotated and the model performs at about 74% F-score on cross-validation tests. A detailed analysis based on empirical evidence shows the contribution of various feature sets to performance. Conclusion: Our experiments indicate a relationship between feature window size and the amount of training data and that a combination of surface words, orthographic features and head noun features achieve the best performance among the feature sets tested. © 2004 Elsevier B.V. All rights reserved.

DOI： 10.1016/j.artmed.2004.07.019

Scopus

PubMed

researchmap

researchmap

researchmap

DOI： 10.1016/j.jbi.2004.08.008

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Japanese Society for Bioinformatics  

DOI： 10.11234/gi1990.14.677

CiNii Article

researchmap

Authorship：Lead author  

researchmap

researchmap

researchmap

researchmap

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper proposes a method to describe conceptual meanings of Japanese deverbal nouns for the purpose of analysing Japanese compound nouns. A lexical meaning of each constituent word is very important since it deeply influences a word formation of compound words. Therefore we intend to use the lexical meanings of constituent words to build a Japanese compound noun analyser. As a first step to describe lexical meanings of words, we have established a simple semantic structure based on lexical conceptual structure (we refer to the structure we establish by TLCS1 in this paper). After the analysing 250 deverbal nouns that appeared in technical terms, we found that the only 14 types of TLCSes cover various deverbal nouns. We will explain in this paper the types of TLCS and also show the procedure of assigning TLCSes to deverbals. In order to evaluate the efficacy of TLCS, we made use of the accuracy of simple compound noun analyser for analysing technical terms. The result was very good, which shows that our proposed method is very promising.

Web of Science

researchmap

researchmap

researchmap

researchmap

Publishing type：Research paper (scientific journal)  

This paper proposes a method to describe conceptual meanings of Japanese deverbal nouns for the purpose of analysing Japanese compound nouns. A lexical meaning of each constituent word is very important since it deeply influences a word formation of compound words. Therefore we intend to use the lexical meanings of constituent words to build a Japanese compound noun analyzer. As a first step to describe lexical meanings of words, we have established a simple semantic structure based on lexical conceptual structure (we refer to the structure we establish by TLCS in this paper). After the analysing 250 deverbal nouns that appeared in technical terms, we found that the only 14 types of TLCSes cover various deverbal nouns. We will explain in this paper the types of TLCS and also show the procedure of assigning TLCSes to deverbals. In order to evaluate the efficacy of TLCS, we made use of the accuracy of simple compound noun analyser for analyzing technical terms. The result was very good, which shows that our proposed method is very promising.

DOI： 10.1109/ICSMC.2001.973032

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

This article provides basic background information on the articles included in this special issue on Japanese term extraction, by (i) clarifying the basic background of research into automatic term recognition, (ii) explaining briefl y the ‘contest’-style workshop we organised in 1999, and (iii) briefl y summarising the ATR methodologies proposed in the articles, and positioning their ideas, philosophies and methodologies within ATR from a unifi ed perspective. Through this information, we intend to consolidate the contributions of the NTCIR TMREC workshop, and, hopefully, clarify a basic framework for discussion which different researchers can use to constructively communicate with each other about automatic term extraction and beyond. © 2000 John Benjamins Publishing Co.

DOI： 10.1075/term.6.2.03kag

Scopus

researchmap

researchmap

researchmap

researchmap

Publisher：National Center for Science Information Systems (NACSIS)  

CiNii Article

researchmap

Other Link： http://dblp.uni-trier.de/db/conf/ntcir/ntcir1999.html#conf/ntcir/KageuraYTKTYO99

Publisher：National Center for Science Information Systems (NACSIS)  

CiNii Article

researchmap

Other Link： http://dblp.uni-trier.de/db/conf/ntcir/ntcir1999.html#conf/ntcir/KageuraYTYTK99

Publisher：National Center for Science Information Systems (NACSIS)  

researchmap

Other Link： http://dblp.uni-trier.de/db/conf/ntcir/ntcir1999.html#conf/ntcir/TakeuchiYKK99

Publisher：National Center for Science Information Systems (NACSIS)  

researchmap

Other Link： http://dblp.uni-trier.de/db/conf/ntcir/ntcir1999.html#conf/ntcir/KoyamaYTK99

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Internal/External technical report, pre-print, etc.  

researchmap

Language：Japanese   Publishing type：Internal/External technical report, pre-print, etc.  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Internal/External technical report, pre-print, etc.  

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Corresponding author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Language：English   Publishing type：Research paper, summary (international conference)  

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Authorship：Corresponding author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (trade magazine, newspaper, online media)  

researchmap

Authorship：Lead author   Language：Japanese   Publishing type：Research paper, summary (national, other academic conference)  

J-GLOBAL

researchmap

Language：English   Publishing type：Book review, literature introduction, etc.   Publisher：John Benjamins Publishing Company  

DOI： 10.1075/term.00010.dro

Scopus

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

日本語用語抽出にあたっては形態素解析器および形態素辞書が必要となるが，実際に専門分野の文書を既存の形態素解析器と形態素辞書を用いて解析した場合，解析精度の制約により，用語抽出性能を低下させる傾向がある．一方で解析誤りの中には，系統的な誤りと考えられるものがあり，さらにはその本来の結果がどのようなものであるのかを推定できる場合もある．これらの誤りについて解析結果を事後的に修正した上で，その結果から用語抽出を行うことにより，抽出性能を向上させることが期待できる。今回の報告では，情報処理分野の文書を解析する際に発生する系統的な誤りパタンがいくつか存在することをを明かにした上で，誤りを修正した結果から用語抽出を行うことにより，用語抽出性能が向上することを報告する．

CiNii Article

CiNii Books

researchmap

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

多くの研究分野において用語の整理と管理の重要性が指摘されながら、実際には十分な対応がなされているとは言い難い。この理由として、そもそも用語を管理する枠組み自体が十分に整備されておらず、また実際の用語管理を継続的に遂行していくにあたっては多大な人的資源を必要とすることなどが挙げられる。本研究では、適切に選定された研究領域ごとに、用語管理を総合的に行うデータベースを構築した上で、用語管理に関わる労力を低減する目的で、用語候補自動抽出機能を利用した、用語登録支援機能を備えたシステムを開発する試みについて発表する。

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper we construct a verb-adjectival verb dictionary to deal with paraphrase relations between verbs and adjectival verbs. This is because we found that some adjectival verbs can express completion state of some verbs ; e.g. an adjectival verb 'healthy' can express the completion state of 'He is cured'. For this purpose, we elaborate a verb-adjectival verb dictionary in which we collect adjectival verbs that can express completion states of verbs. To check the quality of our dictionary, we build a paraphrase system that inputs a verb phrase and then outputs expressions using adjectival verbs. From this construction, we found several phrasal patterns to paraphrase verb expressions. In preliminary paraphrase experiments, we evaluate the quality of our proposed dictionary and paraphrasing system.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

This paper presents Japanese deverbal compound analyzer which disambiguates senses of deverbal nouns and identify their arguments based on Thesaurus of Predicate Argument Structure. In Japanese there appear a lot of complex compound nouns such as 'koukuuki-tsuiraku-jiko-boushi-sisutemu' (airplane-crash-incident-prevention-system) i.e., a system of prevention of airplane crash. The proposed system identifies deverbal nouns such as 'prevention' and 'crash', and idenfies its arguments 'incident' and 'airplane'. To identify senses of deverbal nouns and we utilize example sentences in Japanese Verb Thesaurus as well as manually constructed noun category. The preliminary evaluation of deverbal-sense disambiguation shows that the proposed noun category is effective.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Identifying idioms in sentences is essential to structure text, especially to identify verb semantics. This task is, however, difficult because an idiom has not only different semantics by relation between a verb and its modifier nouns, but also different paraphrases. To solve this, we have constructed idiom identification system using grammatical rules based on user-editable dictionaries. This system can detect idioms exhaustively, absorbing morphological variants such as replacement of particles or transformation of idiom construction caused by insertion of words. It can also cooperate with Argument Structure Annotator to identify verb semantics, dealing with idiom paraphrases and adopting semantic concept on Thesaurus of Predicate Argument Structure. Furthermore, it allows users easily to register idiom rules in the user-editable dictionaries in XML without word segmentation nor POS information. In this paper, we propose specifications of the user-editable dictionaries and an exhaustive detection method to absorb morphological variants. The experimental results of idiom identification task reveal that the proposal method works well and is effective for semantic disambiguation of idioms.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

We compare the term extraction system to extract technical terms automatically in texts. The methods of comparative target are based on statistical-learning-based model and rule-based model. In comparison, since the text data identified term on infection exists, we use this data as correct answer data. In statistical-learning-based model, we build the term extraction system by learning by CRF based on correct answer data. And in rule-based model, we use the extraction system using SRL as a rule-based word extraction language. We experimented in term extraction, and showed that it is good to perform term extraction by statical-learning based model when there are many correct answer data, and by rule-based model when texts depend on some fields.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, we propose a method to identify semantic role labels utilizing a few correctly label-annotated examples. The similarity between input sentence and correct labeled example sentences is evaluated by manually denned similarity function. This is because a statistical learning model doesn't work well without many correct examlpes. We develop Japanese semantic role label identification system based on the proposal method. Moreover, this system can be added deverbal compound analyzer and idiom identification system. We also report the comparison between the proposal method and CRFs in the experiment.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

This manuscript describes evaluation results of Kernel K-means clustering approach comparing with modified Aizawa's co-clustering approach for verb synonym extraction task. Kernel K-means approach is one of the state-of-the-art vector-based clustering method which can divide vector-spaces with non-linear boundary by incorporating Kernel method. Besides the mathematical framework of Kernel K-means can cover Spectral Graph Clustering. In this manuscript, however, we reveal Aizawa's co-clustering approach overcomes Kernel K-means on the verb synonym extraction task (bi-graph clustering) in Japanese. From this results we discuss that the equivalence between graph-vector space in Kernel K-means approach can be limited, and then Kernel K-means decease their accuracy in our verb synonym extraction.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Argument structure is widely recognized as an interface of mapping from grammatical structure of a sentence to shallow semantic structure. In English several large-scale language resources such as FrameNet, Propbank, and Dorr's LCS are proposed and each of them defines a kind of argument structure and some of them construct annotated corpora. These annotated corpora are very useful to build a statistical annotation system of semantic role labels. While in Japanese EDR provided a large-scale annotated corpus of semantic role labels; however the annotated sentences are not collected on the basis of verbs, thus it is hard to utilize the annotated corpus as a training corpus of statistical semantic role label system. Thus we propose another annotation corpus of argument structure on the basis of the Japanese Verb Thesaurus which is provided in previous work. Currently we annotated 1483 sentences for 120 verbs. In this manuscript we confirm that the problem issues of argument structure annotation, current annotation scheme, development of tool and quality of annotated corpus.

CiNii Article

CiNii Books

researchmap

Language：Japanese  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In the previous work we show that a co-clustering approach is suitable for verb synonym extraction because of polysemy of both noun and verb. The proposed approach is promising, however, it has not succeeded in initial graphs of verb-noun pairs. Thus in this paper we present a modified co-clustering approach with recursively extracting other possible verb gourps from the same verb-noun pairs; this enables us to actively obtain other meanings of each verb. The experimental results of verb synonym extraction from Japanese news paper corpus and balanced corpus show that the modified approach outperformed the previous one in precision and recall rates.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper we propose a noun clustering approach on the basis of CBC proposed by Pantel. CBC is a clustering approach that carefully extracts clusters by finding sub-clusters regarded as committees with the same meanings, and try to extract unknown clusters from the remaining elements. In preliminary experiments of Japanese noun clustering, however, we found that CBC does not work well at the measurement of basic similarity between words with context vectors and scoring method that decides to merge sub-clusters. To these problems in this paper we propose to apply Jensen-Shannon formula as a measurement and a new scoring method. In the experimental results of constructing sub-clusters of Japanese nouns from a new paper article we will show that our proposed approaches overcome the approaches in CBC at the clustering accuracy.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We can assume that two different document sets may show similar morpheme occurrence patterns, if the sets both discuss about similar topics with similar discussion manners. In this paper, the authors show the occurrence patterns of morphemes really indicates the similarity of the sets. The authors also show the difference of the patterns in both sets indicate the difference of topics or discussion manner between the sets. The authors also show how to find key morphemes that indicate the similarity or difference of the sets.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this manuscript we discuss how we make design policy of semantic role labels and predicate argument structure. In previous work of construction of language resources such as FrameNet and VerbNet semantic role labels are practically proposed, however, it still leaves room for discussion about SRL's definition, functionality and design policy from a view of how to use in natural language processing. In this research we are annotating SRLs for Japanese news corpus on the basis of a free Verb thesaurus of argument structure. In this manuscript, from the view of event description of natural language, we make sure that 1) the position of predicate argument structure description, 2) semantic role labels contains three functions and 3) SRLs should be annotated on the basis of not what happened but speaker's expression.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

Term classification associate to research sub-domain is an important approach for systematized classification of term candidates extracted from text corpora. The authors, have developed a method which identify some of the important research sub-domains in research abstract corpora. The authors also proved that relatively frequent term candidates extracted from the corpus can be related to identified sub-domains.

CiNii Article

CiNii Books

researchmap

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

Metaphor expression, especially consisting of a noun and verb, is irregular expression thus it is not easy to be dealt with in natural language applications such as machine translation, summarization, and document understanding. We believe it is important that metaphor expressions should be paraphrased into normalized basic expressions in order to extract a fact embedded in the expression: for example a metaphor expression "the wind plays" can be paraphrased into a basic expression "the wind blows". In order to realize this task, in this paper we propose a verb thesaurus organized on the basis of semantic coherency for verbs. From the work of manual construction of a thesaurus on 50 verbs we discuss the possibility of our approach.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

This paper presents our ongoing research for clustering Japanese verbs for constructing Japanese verb lexicon which is founded on the theory of lexical conceptual structure (LCS). The key issue of this research is how to extract a core cluster of Japanese verbs with a highly relating cluster of nouns because not only verbs but also nouns are polysemouns words. In this paper we applied an approach of co-clustering on the basis of graph sctructure into clustering task of verbs and nouns, and present experimental results on Japanese Verb-Case-Noun data from both large Web corpus and Maichini news paper corpus from 1991 to 1998.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We introduce a method for structurizing term candidates extracted from a Japanese domain corpus, based on nesting relations between the candidates. Prom the nesting relations, we can infer hypernym-hyponym relations and related term relations. Arranging both relations separately, we can get clearer hierarchical relations.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

We introduce a method for structurizing term candidates extracted from a Japanese domain corpus, based on nesting relations between the candidates. Prom the nesting relations, we can infer hypernym-hyponym relations and related term relations. Arranging both relations separately, we can get clearer hierarchical relations.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

Term extraction is one of the most important application of natural language processing technologies. Statistic criteria are widely adopted to evaluate the termhood of the extracted candidates. However, it is difficult to evaluate the termhood of less frequent candidates. In this study we propose a method for Japanese composite term extraction in which unproper morpheme patterns are el iminated. Using the new method, high precision of term extraction can be attained for Japanese composite terms.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

In this paper, we propose a clustering method based on structure of compound nouns. Japanese compound nouns usually provide concrete concepts, thus we can assume that compound nouns must be a good clue to index documents for Web document clustering. The other benefit of using compound nouns is their compositional structure that can be divided into sub concepts. This indicates that we can construct hierarchical Web document clustering based on their structure. From the practical experiments of constructing hierarchical clusters on the results of a Web search engine, we succeeded in showing that the clearly clusterized Web documents with understandable hierarchical indexes of compound nouns.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

We present a rule-based model for semantic role labeling because semantic role labels should be assigned by means of semantic relationships between nouns atid verbs. The purpose of this research is to reveal the restritional and preference rules between nouns and verbs. Our model uses the EDR thesaurus as semantics of nouns, and a lexical conceptual structure dictionary as verb semantics. Experimental results show that dealing with systematic nominal categorization is the key to improve the accuracy of semantic role labeling.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

This paper presents our ongoing research for constructing a Japanese verb lexicon which is founded on the theory of lexical conceptual structure (LCS). LCS provides a framework for capturing the relation between syntactic behaviors of lexical items and their semantic properties, which is useful for a range of NLP tasks including translation, paraphrasing and summarization. We discuss design issues involved in LCS dictionary development, and present an overview of the current specification of the lexicon, which is designed to allow successive future refinements.

CiNii Article

CiNii Books

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, we propose a pattern-based term extraction approach for Japanese, applying ACABIT system originally developed for French. The proposed approach evaluates termhood using morphological patterns of basic terms and term variants. After extracting term candidates, ACABIT system filters out non-terms from the candidates based on log-likelihood. This approach is suitable for Japanese term extraction because most of Japanese terms are compound nouns or simple phrasal patterns.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

In this paper we propose the method to analyze relationships between constituents of Japanese compound nouns, especially the dative case relationship in deverbal noun compounds. The kinds of relations between noun-deverbal noun compounds can be roughly classified into two: the one is that the noun becomes an argument of head verb, and the other is that the noun becomes an adjunct of head verb. The relation of the dative case we take up is a kind of argument relation but it has not been investigated carefully. We try to analyze these relationships and to make sure how we can explain those linguistic phenomena based on lexical semantics theory. We introduce lexical conceptual structure (LOS) as a lexical semantic expression and show the possibility to build compound noun analyzer based on LCS to dead with the dative case relations.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper we propose the method to analyze relationships between constituents of Japanese compound nouns, especially the dative case relationship in deverbal noun compounds. The kinds of relations between noun-deverbal noun compounds can be roughly classified into two: the one is that the noun becomes an argument of head verb, and the other is that the noun becomes an adjunct of head verb. The relation of the dative case we take up is a kind of argument relation but it has not been investigated carefully. We try to analyze these relationships and to make sure how we can explain those linguistic phenomena based on lexical semantics theory. We introduce lexical conceptual structure (LCS) as a lexical semantic expression and show the possibility to build compound noun analyzer based on LCS to dead with the dative case relations.

CiNii Article

CiNii Books

researchmap

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

The aim of our research is to build a Japanese compound analyzer using grammatical constraint. To analyze the compounds which have verbs in their heads is to reveal the relationship between incorporated nouns and the head verbal nouns. The kind of relationship is adjunctive and internal argument of a head verb. In our previous research, we categorize verbs based on their argument structure and lexical conceptual structure (LCS), but it is necessary to categorize nouns to reveal the relationship between a noun and a head verb in compounds. Therefore we propose the method to categorize nouns based on argument structure and LCS of head verbs in this paper. In the case of categorization, we also use qualia structure of generative lexicon. After the assumptions of restriction about categorized nouns is constructed, we attempt the experiment of analyzing technical terms (103 words) in the domain of information processing. Finally we show the experimental results that 92 words (89%) can be solved using our restriction and discuss the limitation of our model.

CiNii Article

CiNii Books

researchmap

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

It is difficult problem to know how to prepare lexical knowledge in order to analyze Japanese compound words. In this paper, we analyze technical terms that consist of two verbal nouns by means of the rich lexical knowledge that are a level ordering assumption, an argument structure and a lexical conceptual structure used on morphology or syntax. Our analysis of compound words aims to reveal the relations between the constituent words of compound words. From the results, our method can explain some parts of phenomena of the impossible compound words and the conversion of verbal noun into noun in compound words. The lexical knowledge we have assumed is aimed at describing the details of verbal nouns. While analyzing verbal nouns, we also clarify how to describe the lexical knowledge of noun words.

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：National Institute of Informatics  

Grammatical information for constituent elements of nominal compounds are needed for analyzing nominal compounds, especially those of technical fields. In this research we analyzed the grammatical categories suitable for nominal compound analysis, as it is difficult to apply conventional grammatical categories defined with respect to words in sentences to the analysis of nominal compounds.

CiNii Article

CiNii Books

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Country：Japan

researchmap