研究者詳細 - 吉野　孝

講演・口頭発表等 - 吉野　孝

分割表示545 件中 381 - 400 件目／全件表示 >>

福島拓, 吉野孝, 重野亜久里

研究報告グループウェアとネットワークサービス（GN） 2009年01月15日一般社団法人情報処理学会

現在，在日外国人数は年々増加しており，多言語によるコミュニケーションの機会は増加している．コミュニケーションを行う際，言語の違いは大きな障壁となる．また，医療の分野では医療従事者と患者間での正確性の高い意思の疎通が重要である．このため，支援を行う際に正確性の確保が難しい機械翻訳を用いることは適切ではない．そこで我々は，多言語用例対訳を共有する TackPad の開発を行っている．今回， TackPad の用例対訳群の評価機能の実現とその評価を行った．本稿では，試用実験の結果から次の知見を得た． (1) 評価者と評価閲覧者間での意思の疎通を行うため，評価内容を明確にする必要がある． (2) 「おすすめ度」という言葉を用いると，評価されることに対する抵抗感が減る可能性がある．Recently, the number of foreign residents in Japan is increasing. Consequently, the opportunityof communication among people whose native language are different increases. When people communicate with each other, the difference of the language is a huge barrier. Moreover, it is important to communicate reliably between medical workers and foreign patients in the medical field. Because it is difficult for machine translation to guarantee accuracy, it is not appropriate to use in the medical field. Therefore, we have developed a multilingual parallel-text sharing system, called TackPad. This paper presents the development and evaluation of the parallel-text evaluation function of TackPad. From the results of two experiments, we obtained the following findings. (1) To communicate smoothly between an evaluator and users, we have to specify an evaluation index. (2) The evaluation expression "Recommendation" can relax the uncomfortable by evaluation for evaluator.

A web-based multilingual parallel corpus collection system for the medical field

Takashi Yoshino, Taku Fukushima, Mai Miyabe, Aguri Shigeno

Proceedings of the 2009 ACM SIGCHI International Workshop on Intercultural Collaboration, IWIC'09 2009年

We have developed a Web-based multilingual parallel corpus collection system named TackPad for medical workers and foreign patients. The number of foreign visitors and residents in Japan is increasing. However, it is not sufficient to provide linguistic support via medical translators to people who do not understand Japanese. In particular, poor communication may lead to medical errors. Moreover, there are limitations to a medical translator's support. It is difficult to use machine translations in the medical field because of their inaccuracy. A system that uses parallel corpora with guaranteed accuracy appears to be useful. However, a huge amount of parallel corpora related to specialty fields and in several languages is required in the medical field. It would be difficult for a limited number of people to accumulate a sufficient amount of parallel corpora. Therefore, we have specially designed a support system named TackPad for the medical field
this system can accurately register parallel corpora and supply this data to other systems. TackPad adopts a Web 2.0 approach and can be used to tap people's capabilities for knowledge accumulation. The purpose of TackPad is to collect parallel corpora
other systems can use the corpus via the Internet. We have obtained various responses from the result of the trial experiment. The experiment has demonstrated the practicality of the system, for example, people of various age groups and from various occupations can actually use the system.

Accuracy evaluation of sentences translated to intermediate language in back translation

Mai Miyabe, Takashi Yoshino

ACM International Conference Proceeding Series 2009年

The back-translation method is used to check the accuracy of a sentence translated to a native language. We believe that there exits a positive correlation between the accuracy of sentences translated to an intermediate language and that of back-translated sentences. However, this has not yet been verified. However, some back-translated sentences have high accuracy even if the translated sentence is inaccurate. Therefore, we have to verify the correlation between the accuracy of sentences translated to an intermediate language and that of back-translated sentences. We have evaluated the accuracy of back-translated sentences and that of sentences translated to an intermediate language to establish the correlation between the two accuracies. We have obtained the following results: (1) There exists a positive correlation between the accuracy of sentences translated to an intermediate language and that of back-translated sentences. (2) The occurrence rate of an accuracy mismatch case, wherein a back-translated sentence is accurate but the translated sentence is inaccurate, is less than or equal to 0.5%. (3) Back-translation can be used to check the accuracy of a translated sentence. Copyright 2009 ACM.

Design of Face-to-Face Multilingual Communication Environment for Illiterate People

Mai Miyabe, Takashi Yoshino

INTERNATIONALIZATION, DESIGN AND GLOBAL DEVELOPMENT, PROCEEDINGS 2009年 SPRINGER-VERLAG BERLIN

In the medical field, a serious problem exists with regard to communication between hospital staffs and patients. Currently, although a medical translator accompanies a patient to medical care facilities, round-the-clock or emergency support is difficult to provide due to increasing requests. The medical field has high expectations from information technology. Therefore, we have developed a support system for multilingual medical reception termed M-3. We have installed Our system in the Kyoto City Hospital in Japan. However, we found that our system cannot provide support to illiterate people. If an illiterate person and another person speak different languages, it is difficult the other person to communicate face to face with the illiterate person while explaining the meaning of texts shown on the display of the support system. This is one of the problems specific to the multilingual communication. There is a need to solve this problem. Therefore, we have developed a method to provide support to illiterate people engaging in multilingual face-to-face communication. We use a text-to-speech function implemented using a selector switch to provide support to illiterate people in performing operations using a touch screen. We performed an experiment to examine the effect of the proposed method. The results of the experiment are as follows. (1) From the results of the questionnaire, we find that the subjects are able to operate the selector switch easily. Therefore, we conclude that the method using the selector switch has little effect on the operation of the system. (2) Retrieval time using the text-to-speech function is five times that using the normal operation. We need to consider a structure that can retrieve the required information easily if many readings of texts are required.

A WEB-BASED MULTILINGUAL UTTERANCE COLLECTION SYSTEM FOR THE MEDICAL FIELD

Takashi Yoshino, Taku Fukushima, Ryuichi Nisimura

WEBIST 2009: PROCEEDINGS OF THE FIFTH INTERNATIONAL CONFERENCE ON WEB INFORMATION SYSTEMS AND TECHNOLOGIES 2009年 INSTICC-INST SYST TECHNOLOGIES INFORMATION CONTROL & COMMUNICATION

We have developed a web-based multilingual utterance collection system, named OTOCKER, for the medical field. The purpose of OTOCKER is to act as a voice data collection platform for intercultural communication. Although speech synthesis systems have improved significantly, fluent and smooth speech synthesis is still a problem. Currently, it is difficult to synthesize speech in different languages. In the medical field, in particular, it is important that the intended meaning of spoken words be conveyed effectively along with the different nuances. Therefore, we use an utterance collection system that collects peoplefs voices directly. The limitations of this system are (1) an insufficient number of correct sentences pertaining to the medical field and (2) that easy participation in both voice record and collection are difficult. We can solve the first problem by using a system that collects parallel medical texts. The second problem can be solved by using w3voice - web-based voice-recording system. This system can run only on a web browser. This paper presents the design of OTOCKER, its prototype, and the results of its trial.

Effects of undertaking translation repair using back translation

Mai Miyabe, Takashi Yoshino, Tomohiro Shigenobu

Proceedings of the 2009 ACM SIGCHI International Workshop on Intercultural Collaboration, IWIC'09 2009年

Translation repair plays an important role in intercultural communication that involves machine translation. It can be used to create messages that have very few translation mistakes. The accuracy of the results of translation repair whenan original sentence is rewritten has not yet been evaluated. The improvement brought about by translation repair has to be demonstrated in order to apply translation repair to multilingual communication. Therefore, we have evaluated the translation repair of Japanese-English, Japanese-Chinese, andJapanese-Korean translations using back translation. We have used test sentences with a character count ranging from 15 to 32. On the basis of these evaluation experiments, we have estimated the accuracy and the cost of translation repair. (1) After nearly six rounds of translation repair work in three languages, the average translation accuracy of the sentencesused in the experiment was improved (the meaning of the translated sentences was almost the same as that of the original sentences). In the experiment, 65% of the sentences were improved to the level of highly accurate. Moreover, 99% ofthe sentences were improved to the level of moderately accurate. (2) The cost of repairing a sentence depended on the number of translation-difficult words or phrases that were contained in the sentence. When the quality of a translation was low, finding the word or phrase to be modified was a difficult task. Thus, the cost corresponded to the quality of the translation. Copyright 2009 ACM.

BunBunMovie: Scenario visualizing system based on 3-D character

Tomoya Matsuo, Takashi Yoshino

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2009年

There are many text-based contents, such as novels and script. Those contents have only scenario, and lack visual information. The purpose of this research is to provide visualizing environment that can visualize text-based contents easily. Moreover, such environment can also provide the opportunity to get pleasure out of scenario. It is necessary for visualizing scenario to make various motions of characters and to depict various situations. Therefore we propose motion assortment function to make various motions of characters. The function uses a Japanese dictionary and a thesaurus search. We also propose associated image display function that uses an image search to depict various situations. From the experiments about the motion assortment function, we show that the proposal method can assort some motions. From the experiments of subjective assessment, we found that some subjects inclined to use such easy visualizing environment. © 2009 Springer Berlin Heidelberg.

翻訳リペア支援のための Web 日本語Nグラムを用いた類義語フィルタリング

宮部真衣, 吉野孝

電子情報通信学会技術研究報告. AI, 人工知能と知識処理 2008年11月20日一般社団法人電子情報通信学会

機械翻訳を用いたコミュニヶーションにおいて,翻訳リペアは不適切な翻訳箇所を減少させるための方法として重要な役割を果たす.翻訳リペア作業はユーザへの負担が大きいため,修正の必要な単語の類義語や関連語を提示することによる言い換え作業の支援が必要である.しかし,提示数が多い場合,適切なものを選び出すことは容易ではないと考えられる.本稿では,より適切な言い換え候補の抽出のためにWeb日本語Nグラムを用いたフィルタリングを提案する.また,2-gramおよび3-gramのデータを利用し,前方品詞2-gram,後方品詞2-gram,3-gramの3種類の単語の組み合わせによるフィルタリング実験を行い,以下の知見を得た.(1)Web日本語Nグラムを用いたフィルタリングにより,90%前後の単語について,言い換え候補を7語未満に絞り込むことができており,多数の候補を絞り込むことができる可能性がある.(2)閾値を0とした場合,3-gramによるフィルタリングにおいて抽出失敗率が最も高く(34.4%),後方品詞2-gramが最も低く(15.9%)なった.また,除外失敗率については前方品詞2-gramが最も高く(52.7%),3-gramが最も低く(28.6%)となった.(3)複数品詞により構成される言い換え候補については形態素解析を行い,言い換え候補の構成品詞数に応じて利用するNグラムデータを変更することで,抽出失敗を減少できる可能性がある.

多言語医療受付支援システムの構築と医療機関への導入

宮部真衣, 吉野孝, 重野亜久里

情報処理学会研究報告デジタルドキュメント（DD） 2008年11月20日一般社団法人情報処理学会

医療の現場では，外国人診療時における患者との対話に大きな課題を抱えている．現在は，医療通訳者同行による対応を行っているものの，その需要は急速に増大しており， 24 時間対応や緊急時対応などが困難である．情報技術への期待が大きいものの，長期的な利用可能性を持つ実用的なシステムの実現・導入には至っていない．理由としては，（1）医療分野では，極めて高い翻訳精度が要求されており，機械翻訳技術による支援は難しい，（2）異なる言語を用いる利用者間の対面同期環境における対話は，その状況の特殊性からほとんど検討されておらず，さらに，病院における即時性の高さに対応できていない上記の問題を解決する多言語医療受付支援システム M 3 の構築を行った． M 3 では，医療分野で利用可能な翻訳精度を実現するために，用例対訳を用いる．対面同期環境における多言語対話のためのインタフェースとして，役割に応じたインタフェースおよびフローチャート型情報提供機能を提案する．これらの対応により， M 3 は実際の中規模病院への導入を実現した．In the medical field, a serious problem exists with regard to communications between hospital staff and patients. Currently, although a medical translator accompanies a patient to medical care facilities, round-theclock or emergency support is difficult to provide due to the increasing demand for it. The medical field has high expectations of information technology. However, a useful system has yet to be developed and introduced in the medical field for practical use. The reasons for this are as follows. (1) In particular, the medical field requires highly accurate translations that cannot be achieved by the machine translation technology currently in use. (2) Face-to-face conversations between people speaking different languages are not feasible due to the specific nature of the communications in the medical field. Therefore, multilingual communication cannot respond to situations involving high immediacy. In this paper, we developed a multilingual communication support system, named M3. M3 uses parallel texts to achieve high accuracy in communications between people who speak different languages. Moreover, M3 employs a user-friendly interface and contains a function that provides information using a flowchart. As a consequence, we has introduced M3 into a medium scale hospital.

多言語医療受付支援システムの構築と医療機関への導入

宮部真衣, 吉野孝, 重野亜久里

電子情報通信学会技術研究報告. AI, 人工知能と知識処理 2008年11月20日一般社団法人電子情報通信学会

医療の現場では,外国人診療時における患者との対話に大きな課題を抱えている.現在は,医療通訳者同行による対応を行っているものの,その需要は急速に増大しており,24時間対応や緊急時対応などが困難である.情報技術への期待が大きいものの,長期的な利用可能性を持つ実用的なシステムの実現・導入には至っていない.理由としては,(1)医療分野では,極めて高い翻訳精度が要求されており,機械翻訳技術による支援は難しい,(2)異なる言語を用いる利用者間の対面同期環境における対話は,その状況の特殊性からほとんど検討されておらず,さらに,病院における即時性の高さに対応できていない.上記の問題を解決する多言語医療受付支援システムM^3の構築を行った.M^3では,医療分野で利用可能な翻訳精度を実現するために,用例対訳を用いる.対面同期環境における多言語対話のためのインタフェースとして,役割に応じたインタフェースおよびフローチャート型情報提供機能を提案する.これらの対応により,M^3は実際の中規模病院への導入を実現した.

翻訳リペア支援のためのWeb日本語Nグラムを用いた類義語フィルタリング

宮部真衣, 吉野孝

情報処理学会研究報告デジタルドキュメント（DD） 2008年11月20日一般社団法人情報処理学会

機械翻訳を用いたコミュニケーションにおいて，翻訳リペアは不適切な翻訳箇所を減少させるための方法として重要な役割を果たす．翻訳リペア作業はユーザへの負担が大きいため，修正の必要な単語の類義語や関連語を提示することによる言い換え作業の支援が必要である．しかし，提示数が多い場合，適切なものを選び出すことは容易ではないと考えられる．本稿では，より適切な言い換え候補の抽出のために Web 日本語Ｎグラムを用いたフィルタリングを提案する．また， 2－gram および 3-gram のデータを利用し，前方品詞 2-gram ，後方品詞 2-gram， 3－gram の 3 種類の単語の組み合わせによるフィルタリング実験を行い以下の知見を得た． (1) Web 日本語Ｎグラムを用いたフィルタリングにより， 90％前後の単語について，言い換え候補を 7 語未満に絞り込むことができており，多数の候補を絞り込むことができる可能性がある． (2) 閾値を 0 とした場合， 3－gram によるフィルタリングにおいて抽出失敗率が最も高く（34.4％) ，後方品詞 2-gram が最も低く（159％）なった．また，除外失敗率については前方品詞 2-gram が最も高く（52.7%) ， 3－gram が最も低く（28.6%）となった． (3) 複数品詞により構成される言い換え候補については形態素解析を行い言い換え候補の構成品詞数に応じて利用するＮグラムデータを変更することで，抽出失敗を減少できる可能性がある．Translation repair has an important role in creating text that has a minimum amount of improper translations during communications using machine translation. However, deciding the repair strategy to be used is a heavy burden on the user, and a mechanism to reduce the user burden is required. Thus, we will investigate methods to support paraphrasing by providing synonyms and related words for terms that need repair. In this paper, we propose a filter based on Web Japanese N-grams to extract more appropriate paraphrase candidates. In addition, we carried out an experiment using the proposed method, and found the following results. (1) By using a filter based on Web Japanese N-grams, paraphrase candidates were reduced to less than 7 terms in more than 90% of the words. There is a possibility that a large number of candidates can be narrowed down to a few terms. (2) The maximum rate of extraction failure was 34.4% in 3-gram, and the minimum of that was 15.9% in backward part of speech 2-gram at threshold 0. The maximum rate of elimination failure was 52.7% in front part of speech 2-gram, and the minimum of that was 28.6% in 3-gram. (3) Extraction failure is reduced by change to n-gram data corresponding to the number of parts of speech in each paraphrase candidate.

対面討論における非母語話者支援システムPaneLiveの開発

福島拓, 吉野孝, 喜多千草

ワークショップ2008 （GN Workshop 2008) 論文集 2008年11月06日

パネル討論：アプリケーション開発における音声認識

西村竜一, 中野鐵兵, 栗原一貴, 中臺一博, 吉野孝

情報処理学会研究報告音声言語情報処理（SLP） 2008年10月17日一般社団法人情報処理学会

三次元仮想空間における多言語チャットコミュニケーション実験

吉野孝, 池信克也

情報処理学会研究報告グループウェアとネットワークサービス（GN） 2008年09月18日一般社団法人情報処理学会

三次元仮想空間の一つである Second Life 上で固定型および携帯型の 2 種類の多言語コミュニケーション支援システムを構築し，多言語コミュニケーション実験を Second Life 上で実施した．実験の結果，下記の知見を得た． (1) Second Life 上に機械翻訳を用いたチャットコミュニケーションを単純に導入しても，チャット特有の表現などの問題のため，コミュニケーションの実現は難しい． (2) Second Life 上の利用者に機械翻訳を利用していることを明示することで，機械翻訳を用いたチャットを行うことができる可能性がある．We have developed a multilingual chat communication support system on Second Life. The Second Life is one of popular 3D online virtual spaces in the world. We carried out a mul-tilingual chat communication experiment on Second Life. FVom the experiments, we found the following findings. (1) When we use a multilingual chat communication support system with ill-advised, it is difficult to communicate with Second Life's people because of typical way of chat. (2) When we express the existence of machine translation, we can support a multilingual chat communication on Second Life.

J-009 言語グリッドを用いた三次元仮想空間における多言語チャットコミュニケーション支援システムの構築(ヒューマンコミュニケーション&インタラクション,一般論文)

池信克也, 吉野孝

情報科学技術フォーラム講演論文集 2008年08月20日 FIT(電子情報通信学会・情報処理学会)運営委員会

J-016 リアルタイム遠隔コミュニケーションにおけるアウェアネス情報の対人許容レスポンス時間への影響(ヒューマンコミュニケーション&インタラクション,一般論文)

宮部真衣, 吉野孝

情報科学技術フォーラム講演論文集 2008年08月20日 FIT(電子情報通信学会・情報処理学会)運営委員会

言語グリッドを用いた三次元仮想空間における多言語講義室の構築

吉野孝, 池信克也, 重信智宏

情報処理学会研究報告コンピュータと教育（CE） 2008年05月10日一般社団法人情報処理学会

三次元仮想空間は，新しいコミュニケーション空間になりうると考えられるが，現実空間と同様の言語障壁が発生している．そこで，三次元仮想空間における多言語コミュニケーション環境の一つとして，多言語講義室の構築を行った．多言語講義室は，言語グリッドを介した機械翻訳を利用することで，利用者自身で専門用語の対訳辞書を構築できる機能を提供する．We think that a virtual space can be a new communication environment. However, there exists a language barrier on Second Life that is a famous 3D online virtual space. Therefore, we have developed a multilingual classroom on the Second Life. The multilingual classroom can provide the function of construction of the dictionary of technical term based on Language Grid.

三次元仮想空間上における多言語チャットコミュニケーション支援

池信克也, 吉野孝

全国大会講演論文集 2008年03月13日

在宅診療現場における同期・非同期対応型情報共有システムの開発とその評価

榎本紗耶香, 吉野孝, 紀平為子, 入江真行

全国大会講演論文集 2008年03月13日

用例読み上げ機能を持つ多言語医療受付対話支援システムM^3

宮部真衣, 吉野孝

全国大会講演論文集 2008年03月13日

　前のページ - 次のページ　

<吉野 孝>

講演・口頭発表等 - 吉野 孝

<吉野　孝>

講演・口頭発表等 - 吉野　孝