会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明专利
    • Dictionary-registering device for natural language processing, registration method of dictionary for natural language processing, and program
    • 用于自然语言处理的字幕注册设备,自然语言处理词典的注册方法和程序
    • JP2004213083A
    • 2004-07-29
    • JP2002378448
    • 2002-12-26
    • Toshiba Corp株式会社東芝
    • KUMANO AKIRAKINOSHITA SATOSHI
    • G06F17/28
    • PROBLEM TO BE SOLVED: To provide a registration device for natural language dictionary which can prevent errors from occurring, when a type of conjugation is selected, and can perform efficient registration work by flexible input operation. SOLUTION: In a GUI, consisting of an input part 1 and an output part 5, if the original form of a phase of a language to be registered is inputted and a selection button for the type of existing conjugational variations is depressed, a conjugational variation module 313 of a conjugation information setting part 31 of a registration processing part 3 generates conjugation, based on the original form and the type of conjugational variations, and displays the conjugation in a conjugational variation display/input column which doubles as a display column and an input column for the result of the conjugational variations. Here, if a user notices an error in conjugation, he or she can correct it by selecting the correct type of conjugation again, but he or she can also directly modify the conjugation displayed in the conjugational variation display/input column. COPYRIGHT: (C)2004,JPO&NCIPI
    • 要解决的问题:提供一种可以防止错误发生的自然语言字典的注册装置,当选择一种共轭时,可以通过灵活的输入操作来执行有效的注册工作。

      解决方案:在由输入部分1和输出部分5组成的GUI中,如果输入要注册的语言的相位的原始形式,并且按压现有共轭变化类型的选择按钮, 注册处理部分3的共轭信息设置部分31的共轭变化模块313基于原始形式和共轭变化的类型产生共轭,并且将共轭显示在共同变化显示/输入列中,该共轭变化显示/输入列兼作显示器 列和用于结合变化结果的输入列。 这里,如果用户注意到共轭错误,则他或她可以通过再次选择正确的共轭类型来进行纠正,但他或她也可以直接修改共轭变异显示/输入列中显示的共轭。 版权所有(C)2004,JPO&NCIPI

    • 2. 发明专利
    • Keyword input support device, keyword input support method and program
    • 关键字输入支持设备,关键字输入支持方法和程序
    • JP2009289202A
    • 2009-12-10
    • JP2008143644
    • 2008-05-30
    • Toshiba Corp株式会社東芝
    • SUZUKI MASARUKINOSHITA SATOSHIUMEKI HIDEONAKANO WATARU
    • G06F17/30
    • G10L15/1822G06F17/30616
    • PROBLEM TO BE SOLVED: To easily input a keyword when conducting an information search or the like. SOLUTION: A keyword input support device includes: a document acquisition unit 101 for acquiring a document having a plurality of component elements including text data; a main element selecting unit 102 for selecting the component element with a large number of characters of the text data among the component elements as a main element showing the central contents of the document; a morphological analysis unit 103 for subjecting the text data included in the main element to morphological analysis, and providing a meaning attribute to each word in the text data subjected to the morphological analysis; a proper representation extracting unit 105 for extracting words with predetermined meaning attributes or parts of speech as proper representations from the words provided with the meaning attributes; a proper representation storage unit 107 for storing the proper representations together with the corresponding meaning attributes; a keyword candidate sorting unit 108 for sorting the proper representations as keyword candidates from the proper representation storage unit 107 according to the meaning attributes; and a keyword candidate presenting and selecting unit 110 for presenting the keyword candidates to the user. COPYRIGHT: (C)2010,JPO&INPIT
    • 要解决的问题:在进行信息搜索等时容易地输入关键字。 解决方案:关键字输入支持设备包括:文档获取单元101,用于获取具有包括文本数据的多个组成元素的文档; 主要元素选择单元102,用于在组成元素中选择具有大量文本数据的字符的组成元素作为表示文档的中心内容的主要元素; 形态分析单元103,用于对包含在主要元素中的文本数据进行形态分析,并且对经过形态分析的文本数据中的每个单词提供含义属性; 适当的表示提取单元105,用于从具有意义属性的单词中提取具有预定含义属性或词性的词作为适当表示; 适当的表示存储单元107,用于存储适当的表示以及相应的含义属性; 关键词候选排序单元108,用于根据意义属性从适当表示存储单元107对适当表示作为关键字候选进行排序; 以及用于向用户呈现关键字候选的关键字候选呈现和选择单元110。 版权所有(C)2010,JPO&INPIT
    • 3. 发明专利
    • Natural language analyzing device
    • 自然语言分析装置
    • JP2003345791A
    • 2003-12-05
    • JP2002150049
    • 2002-05-24
    • Toshiba Corp株式会社東芝
    • HIRAKAWA HIDEKIONO KENJIDEWA TATSUYAKINOSHITA SATOSHI
    • G06F17/27
    • PROBLEM TO BE SOLVED: To provide a natural language analyzing device which provides a sentense analyzing system for allowing a syntax tree being an explanation for the whole kinds of syntax to correspond to a dependent relation tree responding to it concerning the analysis of a natural language, and also which treats probabilities of the whole kinds of dependent relation. SOLUTION: A syntax analysis algorithm for performing analysis by adapting a rewriting rule to an input sentence includes: a means for detecting a category in the configuration element of the rewriting rule when the syntax is analyzed; a means for detecting the partial range of the input sentence dominated by the configuration element; and a means for detecting a main configuration element in correspondence relation with one or a plurality of words in the partial range. COPYRIGHT: (C)2004,JPO
    • 要解决的问题:提供一种自然语言分析装置,其提供了一种声信息分析系统,用于允许语法树是整个语法的解释,以对应于响应于该关系树的分析的依赖关系树 自然语言,也是处理各种依赖关系的概率。 解决方案:一种用于通过将重写规则适应于输入句子进行分析的语法分析算法包括:在分析语法时检测重写规则的配置元素中的类别的装置; 用于检测由配置元素主导的输入句的部分范围的装置; 以及用于检测与部分范围中的一个或多个单词对应关系的主配置元素的装置。 版权所有(C)2004,JPO
    • 4. 发明专利
    • Device, method, program, and system for machine translation
    • 用于机器翻译的设备,方法,程序和系统
    • JP2009075791A
    • 2009-04-09
    • JP2007243195
    • 2007-09-20
    • Toshiba Corp株式会社東芝
    • SUZUKI HIROKAZUKINOSHITA SATOSHI
    • G06F17/28G06F17/30
    • G06F17/2809
    • PROBLEM TO BE SOLVED: To provide a machine translation device for improving the precision of translation. SOLUTION: This machine translation device is provided with a receiving part 101 for receiving a translation request including an input sentence and dictionary information; a text acquisition part 102 for calculating similarity between the input sentence and a text, and for acquiring a text whose similarity is higher than a threshold from a text storage part 121; a dictionary information acquisition part 103 for acquiring the dictionary information having a dictionary information ID corresponding to the acquired text from a dictionary storage part 122; a translation part 104 for determining whether or not a first word in the acquired dictionary information is included in the input sentence, and for, when it is included, translating the first word included in the input sentence into a corresponding second word in the acquired dictionary information; and a storage part 105 for storing the dictionary information included in the translation request in a dictionary storage part 122, and for storing dictionary information ID of the stored dictionary information and the input sentence in a text storage part 121 in association with each other. COPYRIGHT: (C)2009,JPO&INPIT
    • 要解决的问题:提供一种用于提高翻译精度的机器翻译装置。 解决方案:该机器翻译装置设置有用于接收包括输入句子和字典信息的翻译请求的接收部分101; 用于计算输入句子和文本之间的相似度并从文本存储部分121获取相似度高于阈值的文本的文本获取部分102; 字典信息获取部分103,用于从字典存储部分122获取具有与获取的文本相对应的字典信息ID的字典信息; 翻译部分104,用于确定所获取的字典信息中的第一个单词是否包括在输入句子中,并且当包括在内的时候,将包含在输入句子中的第一个单词转换成所获取的词典中相应的第二个单词 信息; 以及存储部分105,用于将包括在翻译请求中的字典信息存储在字典存储部分122中,并且用于将存储的字典信息和输入的句子的字典信息ID存储在文本存储部分121中。 版权所有(C)2009,JPO&INPIT
    • 5. 发明专利
    • Mechanical translation device and method, and mechanical translation program
    • 机械翻译设备和方法以及机械翻译程序
    • JP2005038291A
    • 2005-02-10
    • JP2003276249
    • 2003-07-17
    • Toshiba Corp株式会社東芝
    • KUMANO AKIRASUZUKI HIROKAZUKINOSHITA SATOSHI
    • G06F17/28
    • PROBLEM TO BE SOLVED: To provide a mechanical translation device for converting a word, even if it is not recorded in a translation dictionary, to a right translation. SOLUTION: In the mechanical translation device for translating a first language to a second language, a second language expression extraction part 150 analyzes an existing document described in the second language and collects expressions of the second language. A translation determination part 114 gives a translation to an expression which appears in a translation object document of the first language in reference to second language expression extraction data extracted by the second language expression extraction part 150. COPYRIGHT: (C)2005,JPO&NCIPI
    • 要解决的问题:提供一种机械翻译装置,用于将词语转换成正确的翻译,即使它没有被记录在翻译词典中。 解决方案:在用于将第一语言翻译成第二语言的机械翻译装置中,第二语言表达提取部分150分析以第二语言描述的现有文档并且收集第二语言的表达。 参照由第二语言表达提取部分150提取的第二语言表达提取数据,翻译确定部分114给出出现在第一语言的翻译对象文档中的表达式的翻译。(C)2005, JPO&NCIPI
    • 6. 发明专利
    • Input correction method and device
    • 输入校正方法和设备
    • JPH11282838A
    • 1999-10-15
    • JP8687498
    • 1998-03-31
    • Toshiba Corp株式会社東芝
    • SAKO TAKAYUKIKINOSHITA SATOSHI
    • G06F17/21G06F17/27
    • PROBLEM TO BE SOLVED: To provide an input correction method which can attain smooth input of a document and also can reduce the burden on a user. SOLUTION: A control part 102 sends the analysis result that is received from a character string analysis part 103 to an input correction part 105. The part 105 properly corrects the analysis result as necessary to return this corrected result to the part 102 or to return the received analysis result as it is to the part 102 if no correction is required. The part 102 sends the received analysis result to an output part 106. The analysis of the part 103 is identical with the analysis of a morpheme, for example. In regard to correction of the analysis result, an input error is detected and corrected by referring to the correction processing rule described based on a set of a 1st condition concerning the analysis result, a 2nd condition that should be satisfied by the analysis result which satisfies the 1st condition and the information on the correction to be carried out when the analysis result does not satisfy the 2nd condition, for example.
    • 要解决的问题:提供可以实现文档的平滑输入的输入校正方法,并且还可以减轻用户的负担。 解决方案:控制部分102将从字符串分析部分103接收的分析结果发送到输入校正部分105.部件105根据需要适当地校正分析结果,以将该校正结果返回到部件102或返回 如果不需要校正,则收到分析结果。 部分102将接收到的分析结果发送到输出部分106.例如,部分103的分析与语素的分析相同。 关于分析结果的校正,通过参照基于关于分析结果的第一条件的集合描述的校正处理规则来检测和校正输入错误,满足关于分析结果的分析结果应满足的第二条件 第一条件和当分析结果不满足第二条件时要进行的校正的信息。
    • 7. 发明专利
    • TRANSLATION DEVICE AND DICTIONARY PRIORITY SETTING METHOD
    • JPH0981572A
    • 1997-03-28
    • JP23403095
    • 1995-09-12
    • TOSHIBA CORP
    • YOSHIMURA YUMIKOKINOSHITA SATOSHI
    • G06F17/28
    • PROBLEM TO BE SOLVED: To reduce the labor of words selection in translation and post-editing by taking the result of user's words selection in translation as learning materials to adjust the initial priority designation contents to a priority more adapted to a translation object document without user's knowing. SOLUTION: A first language as the translation object is translated to a second language by a translation part 14 in accordance with classifications and priorities of dictionaries designated as optional dictionaries out of dictionaries preliminarily prepared in a dictionary part 15. In this case, it is discriminated whether the priority change is necessary or not based on selection contents of words selection obtained as the translation result by a priority setting part 12; and if it is necessary, priorities are so adjusted that the dictionary from which words in translation are selected most frequently has the highest priority. Thus, words in translation registered in plural dictionaries are always generated in a translation with optimum priorities by user's words selection in translation, and as the result, the labor of user's words selection in translation and post- editing is reduced.
    • 8. 发明专利
    • DOCUMENT PREPARING DEVICE
    • JPH07200603A
    • 1995-08-04
    • JP35116693
    • 1993-12-28
    • TOSHIBA CORP
    • KINOSHITA SATOSHITAKEDA KIMITOHIRAKAWA HIDEKI
    • G06F17/28
    • PURPOSE:To provide a document preparing device which can effectively support the correcting and editing jobs of the translated sentences. CONSTITUTION:A document preparing device is provided with an input means 1 which inputs the characters and the editing commands, an editing means 3 which refers to the first document written in the first language and also produces or edits the second document written in the second language, a display means 4 which displays at least two documents under reference or production/ editing, and a dictionary storage means 5 which stores a dictionary where at least the translated words of the second language are described against the words written in the first language. The means 5 has a function to refer to the contents of the dictionary against the words contained in the first document and also another function to convert a specific character train inputted to the second document into a word of the second language based on the reference result of the dictionary.
    • 10. 发明专利
    • JPH05252208A
    • 1993-09-28
    • JP4555792
    • 1992-03-03
    • TOSHIBA CORP
    • UENO MOTOHARUDATAKE KENJIKINOSHITA SATOSHIUEHARA KIYOHIKOOKAMOTO TOSHIO
    • H04L9/32H04L9/00H04L9/10H04L9/12H04L12/70H04L12/56
    • PURPOSE:To prevent illegal interception by inputting data to a processing definition program sent in a packet so as to implement the processing and discriminating whether or not the result of processing is able to be sent to a node of its own station side and sending the packet when available. CONSTITUTION:A control section 21 checks the validity of a content of a received packet 14 and permits that a processing definition program of the packet 14 is in operation when the packet is a normal packet. In this case, the control section 21 discriminates the validity of the packet 14 depending whether or not the outgoing node identification symbol of the packet 14, the attribute of the packet of a charging node identification symbol and the content of the processing definition program are proper. When the execution by the packet 14 is finished, the control section 21 discriminates the availability of the transmission depending on the content of a transmission packet 20 stored in a 3rd memory 19 and the content of a state information serve section 17 and when available, the packet 20 is sent to an object node. Thus, the transmission to another node is permitted only when the packet is not illegal and a care about mis-use of data is avoided.