Nec Corporation

Kenji Satoh

Takahiro Ikeda

Yousuke Sakao

Satoshi Nakazawa

Shinichi Ando

Kazunori Muraki

Susumu Akamine

Akitoshi Okumura

Shinichi Doi

Shinichiro Kamei

Kiyoshi Yamabana

Yoshihiro Ikeda, Legal Representative

In a voice recognizing apparatus, a voice recognizing method, and a program for recognizing voice are provided to carry out voice recognition with high precision. When a call receiving unit in which a plurality of telephone numbers for receiving are assigned to respective contents of calls receives a call from a caller, a calling voice signal input device inputs a calling voice signal of the call, and a receiving telephone number input device inputs the receiving telephone number of the call. A voice correction device then carries out voice correction processing for the calling voice signal in accordance with the receiving telephone number and a voice recognizing device executes voice recognition processing for the calling voice signal in accordance with the receiving telephone number.

Voice recognizing apparatus, voice recognizing method, and program for recognizing voice

The question and answer data editing device for editing dialog content to generate question and answer data, includes a detecting unit that detects a part of the dialog content similar to existing question and answer data stored, and a extracting unit that extracts a context in which the dialog content is made from dialog content in the proximity of the similar part detected and registers the context extracted as new question and answer data or as index information of the question and answer data.

Question and answer data editing device, question and answer data editing method and question answer data editing program

[PROBLEMS] To provide a data processing device such as a text mining device capable of extracting characteristic structures properly even in case a plurality of words indicating identical contents or a plurality of words semantically associated are contained in input data. [MEANS FOR SOLVING PROBLEMS] Association node extraction unit () of a text mining device () extracts association nodes containing semantically associated words from a graph obtained as a result of syntax analysis. Association node joint unit () transforms the graph by joint of a part of or a whole of the association nodes. Characteristic structure extraction unit () extracts a characteristic structure from the graph transformed by the association node joint unit.

Data processing device, data processing method, and data processing program

Language analysis meansanalyzes texts read from a text DB, and generates a sentence structure as the analysis result. Similar-structure generation adjustment meansgenerates, from an input of an input device, a determination item for determining whether or not the structures are identical every type of differences between the sentence structures. Similar-structure determination adjustment meansgenerates, from an input of the input device, a determination item for determining whether or not the difference between attribute values is ignored every type of attribute values. Similar-structure generating meansgenerates a similar structure of a partial structure forming the sentence structure obtained by language analysis meansin accordance with the determination item from the similar-structure generation adjustment means, and sets the generated similar structure as an equivalent class of the partial structure on the generation source. Frequent-similar-pattern detection meansignores the attribute value in accordance with the determination item given from the similar-structure determination adjustment means, detects the frequent pattern on the basis of a set of equivalent classes from the similar-structure generating means, and outputs the frequent pattern to an output device

Text mining device, method thereof, and program

With respect to each part at which a word included in a characteristic condition defining a characteristic text set designated by a user through the input device appears in text, the characteristic condition assurance degree calculating unit of the text mining device obtains a reliability of the word from the word reliability storage unit to operate a value of a characteristic condition assurance degree for each text by predetermined operation based on all the obtained reliabilities. The characteristic condition assurance degree calculating unit executes operation such that when a value of each reliability is large, a value of a degree of assurance becomes large. The representative text output unit outputs text whose characteristic condition assurance degree is the highest among texts whose characteristic condition assurance degrees are calculated together with its characteristic condition assurance degree.

Text mining device, text mining method, and text mining program

A translation supporting apparatus which searches out a translation example useful for a translation task from within a translation example database is disclosed. The translation example database stores character strings of a first language and translation results of a second language corresponding to the character strings in a unit of a document. A retrieval request inputting apparatus inputs a translation target sentence. A similarity retrieval apparatus determines, for each translation example, a similarity to the translation target sentence, a similarity to a translation example context which is another translation example having such a predetermined relationship that it is included in the same document and is present within one sentence before or after the translation example, a similarity to a retrieval request context which is another translation target character string having such a predetermined relationship that it is included in the same document as the translation target character string and is present within the range of one sentence before or after the translation target character string, and a similarity between the translation example context and the retrieval request context, and integrates the four similarities. A similar example outputting apparatus refers to the integrated similarities and outputs those translation examples similar to the translation target character string.

TRANSLATION SUPPORTING APPARATUS AND METHOD AND COMPUTER-READABLE RECORDING MEDIUM, WHEREIN A TRANSLATION EXAMPLE USEFUL FOR THE TRANSLATION TASK IS SEARCHED OUT FROM WITHIN A TRANSLATION EXAMPLE DATABASE

A document storage stores a plurality of documents belonging to a certain field. The administrator of a host device uses a classification attribute selector to select the types of attributes suitable for classifying the documents stored in the document storage, from the types of attributes which are stored in a candidate storage and can be used as classification keys, and stores the selected types of attributes in a classification attribute storage. An attribute analyzer analyzes each of documents retrieved by a document retriever to determine which one of the types of attributes stored in the classification attribute storage an attribute element contained in the document belongs to. An attribute classifier classifies each of the retrieved documents into a category according to the type of the attribute element contained in the document based on the determined type of an attribute from the attribute analyzer.

Classification of retrievable documents according to types of attribute elements

A document management system is provided to manage electronic documents in connection with copyrights. Herein, copyright management information is set with respect to a constituent element of a document that a user produces. The constituent element corresponds to a chapter, a paragraph, a sentence, a figure or a table, for example. The copyright management information represents copyright information regarding the use of the document, which is classified into prescribed levels such as a reference level, a copy level and a print level. In addition, it represents prescribed conditions as to whether the use of the document is allowed with payment of fee, with filing of application and with display of copyright or not, for example. Further, the content of the copyright management information can be changed in response to user levels respectively. The copyright management information is connected with information representing the constituent element. When the user requests to use a prescribed portion of another document other than the document that the user is presently producing, a use-allowed part is determined within the prescribed portion on the basis of the copyright management information of the constituent element overlapping with the prescribed portion. So, the user is capable of making a decision whether to actually use the use-allowed part in consideration of the prescribed conditions.

Determining copyright information based on overlapping information between portions of two documents

The present invention discloses a semiconductor device and its manufacture by which damages, such as cracks, generated by the heat of melting of a fuse that is employed for isolating a circuit from the other circuits, can be blocked from propagating into other parts of the semiconductor device. A lower protective film formed on an oxide film provided on a substrate, has a width larger than that of the fuse, and blocks the propagation of damages generated at melting of the fuse. A first insulating film formed on the oxide film so as to cover the lower protective film, has two grooves reaching the lower protective film that is formed so as to surround the fuse. The fuse is formed in the region between the two grooves formed in the first insulating film. A second insulating film is formed on the first insulating film so as to cover the fuse, and has grooves connected to the grooves formed in the first insulating film. A side-face protective film is formed within the grooves formed in the first and second insulating films and blocks the propagation of damages generated at melting of the fuse.

Semiconductor device having protective films surrounding a fuse and method of manufacturing thereof

A system which allows utilization of retrieval terms used in retrievals performed formerly by an outsider to facilitate and accelerate retrieval of documents. A history storage apparatus stores retrieval histories each having information including a group of retrieval items including at least an item of 'retrieving person', retrieval terms and a retrieval result. A history storage section stores, when retrieval of documents is performed, a retrieval history regarding the retrieval into the history storage apparatus. A history retrieval section retrieves, in 'retrieval which is performed utilizing a retrieval history', a retrieval history which satisfies designations of terms of an inputted retrieval item from among the retrieval histories by another retrieving person stored in the retrieval storage apparatus. A retrieval section retrieves documents utilizing retrieval terms of a retrieval history by the other retrieving person detected by the history retrieval section.

Information retrieval system

In order to provide a front-end processor wherein additional functions for retrieving external resources can be easily supplemented optionally without user's complicated environment setting, a front-end processor of the invention, having a main processing section (2) and retrieval sections (4) controlled through the main processing section, comprises: notification means (10) provided in each of the retrieval sections for transmitting initialization information to the main processing section when a file of said each of the retrieval sections is deployed in a main memory area; and initialization means (7) provided in the main processing section for initializing default environments of the front-end processor according to the initialization information transmitted from each of the retrieval sections.

Front-end processor with notification and initialization means

A multistage front end processor system combining a plurality of front end processors serially for executing complicated pre-processing. The plurality of stand-alone front end processors (6-1 to 6-n) are programs such as a penned character recognition program and a kana-kanji conversion program. A definition information storage module (7) may contain definition information on a composite front end processor composed of a plurality of stand-alone front end processors connected serially. A control module (9) references a definition information storage module (7) and, when a composite front end processor to be started is composed, for example, of a first-stage and a second-stage stand-alone front end processors (6-1, 6-2), sends data entered from an input device 1 to the first-stage stand-alone front end processor (6-1), sends its output to the second-stage stand-alone front end processor (6-2), and sends its output to an application program 10.

Multistage front end processor system which uses either a stand-alone

The relationships between words included in an input character string are analyzed. At least one related element conforming to the analyzed relationships is extracted from the input character string. Subject sentences are retrieved from stored documents using the related elements as retrieval keys, and the retrieved subject sentences are displayed on screen. Priority rank is given to the respective subject sentences in accordance with the degrees of coincidence between the respective subject sentences and the related elements. The subject sentences are displayed on screen according to the given priority rank.

Growing community of inventors

Tokyo, Japan

Kenji Satoh