JP2019160134A - Sentence processing device and sentence processing method - Google Patents

Abstract

To provide a sentence processing device and a sentence processing method capable of classifying a vast amount of text into keywords corresponding to news and fashion, summarizing the sentence according to a sentence structure, and obtaining information from a specific viewpoint from a specific point of view.SOLUTION: The sentence processing device inputs a classification tag name for classification and a specification classification tag composed of keywords together with a sentence, and calculates the degree of validity when the specification classification tag is used as the classification tag of the sentence, and when the degree of the sentence with the input classification tag input is equal to or greater than the prescribed threshold, the device assigns a specified classification tag to the sentence (Phase I). In the summary of the sentence, the device analyzes the structure and meaning of the sentence, adds information on the classification in Phase I to the information, and creates a summary sentence (Phase II).SELECTED DRAWING: Figure 3

Description

  The present invention relates to a sentence processing apparatus and a sentence processing method, and more particularly to sentence processing suitable for quickly classifying enormous sentences according to purposes and efficiently obtaining a summary.

  Along with the explosive spread of the Internet, how companies use Web systems to deal with customers is becoming the most important issue. Under such circumstances, the company collects information not only from the information transmission using the Web system but also from the customer's request and writing on the product / service in conjunction with the call center and SNS (Social Networking Service), Identifying improvements and responding immediately to such requests are important points in corporate activities.

  In order for a company to effectively use documents collected from the Internet, a technology for classifying and summarizing documents for each topic by an information processing device is required. For example, Patent Document 1 discloses a method of selecting a question category based on extracted person attributes and topics as candidates and classifying user utterances as answers to questions belonging to the question category. Patent Document 2 discloses a technique for outputting a plurality of summary sentences from input text, evaluating the summary sentences from a plurality of evaluation scales, and extracting the most appropriate summary sentence.

JP 2006-206894 A JP 2006-161967 A

  For example, in a large company that performs diversified management such as a railway operator, the number of documents collected from the Internet reaches a level of hundreds of thousands per year. Such document information should be appropriately and timely processed and utilized in corporate activities.

  In particular, there are cases where sentences related to keywords that are trending on the Internet are extracted, processed in a timely manner, and opinions and correspondence information are required. For example, it is necessary for a railway operator to acquire information from a sentence about keywords related to an accident on a specific route and take appropriate measures.

  The technique described in Patent Document 1 does not consider such a technique for specifying and classifying specific keywords, and cannot classify sentences that deal with news and trends. Further, Patent Document 2 does not disclose a technique for creating a summary regarding a specific viewpoint such as a user's request. For example, it is not considered to improve the customer satisfaction by extracting an improvement point according to the user's request by extracting a summary sentence based on the viewpoint related to the request from the user and dealing with it.

  The purpose of the present invention is to be able to classify a large number of target sentences with keywords that deal with news and trends, summarize sentences according to the sentence structure, and more appropriately from a specific viewpoint than the summarized sentences. An object of the present invention is to provide a sentence processing apparatus and a sentence processing method capable of acquiring information.

  The sentence processing apparatus of the present invention is preferably a sentence processing apparatus that classifies a sentence by adding a classification tag to the sentence, and inputs the sentence and the designated classification tag, and the designated classification tag and the sentence, Calculates the degree of fit, which is an index indicating the validity of the specified classification tag as the classification tag of the sentence, and designates the sentence when the degree of sentence fit with the specified classification tag is greater than the specified threshold A classification tag is added.

In another example of the sentence processing apparatus of the present invention, preferably, a sentence processing apparatus that performs sentence summarization and outputs a summary sentence of the sentence, the sentence with the classification tag and the summary sentence It holds summary performance data that is a pair, performs structural analysis on the single sentence of the input sentence, extracts the meaning related to structural analysis, and for each classification tag of summary performance data, the meaning related to structural analysis When a single sentence is extracted and the similarity between the sentence of the input sentence and the sentence of the summary performance data is equal to or higher than the specified threshold, the summary sentence of the summary performance data is made the summary sentence of the input sentence. It is a thing.
The present invention is also configured as a text processing method in the text processing apparatus.

  According to the present invention, it is possible to classify a large number of target sentences by keywords that deal with news and trends, summarize sentences according to the sentence structure, and more appropriately from a specific viewpoint than the summarized sentences. Information can be acquired.

It is a figure which shows the function structure of a text processing apparatus. It is a figure which shows the hardware software configuration of a text processing apparatus. It is a figure which shows the outline | summary of the characteristic of the text processing of this invention. It is a figure which shows an example of classification | category performance data. It is a figure which shows an example of summary performance data. It is a figure which shows an example of classification | category / summarized data. It is a figure which shows an example of text analysis result data. It is a figure which shows an example of a report creation start screen. It is a figure which shows an example of a text classification / summary result screen. It is a figure which shows an example of a text classification and summary detailed list screen. It is a figure which shows an example of a classification change screen. It is a figure which shows an example of a summary change screen. It is a figure which shows an example of a report display screen. It is a flowchart which shows a text information storage process. It is a flowchart which shows a text analysis process. It is a flowchart which shows a sentence classification | category process. It is a flowchart which shows a text summary process. It is a flowchart which shows a classification | category learning process. It is a flowchart which shows the summary learning process. It is a flowchart which shows report preparation processing. It is a flowchart which shows a semi-automatic classification process.

Hereinafter, an embodiment according to the present invention will be described with reference to FIGS.
First, the configuration of a text processing apparatus according to an embodiment will be described with reference to FIGS. 1 and 2.

  As shown in FIG. 1, the sentence processing apparatus 10 according to the present embodiment includes a sentence analysis unit 110, a sentence classification unit 120, a classification change unit 130, a sentence summary unit 140, a summary change unit 150, and a report creation Part 160.

  The sentence analysis unit 110 is a part that analyzes the structure of a sentence received from the outside and outputs sentence analysis result data (details will be described later). The sentence classification unit 120 is a part that classifies (categorizes) sentences based on the words appearing in the sentences and adds classification tags. The classification changing unit 130 is a part that changes the classification classified by the sentence classification unit 120 and re-adds the classification tag. The sentence summary unit 140 is a part that summarizes sentences and creates the summary sentence. The summary changing unit 150 is a part that changes the summary sentence created by the sentence summarizing unit 140. The report creation unit 160 is a part that outputs a result to the user based on the information that classifies the text and the summarized information.

  The sentence analysis unit 110 includes a word analysis unit 111, a structure analysis unit 112, and a sentence analysis result processing unit 113. The word analysis unit 111 is a part that analyzes a sentence at a word level and divides the sentence into words. The structure analysis unit 112 is a part that analyzes the sentence at the structure level and extracts the meaning. The sentence analysis result processing unit 113 is a part that processes the results of the word analysis unit 111 and the structure analysis unit 112.

  The sentence classification unit 120 includes a classification processing unit 121, a semi-automatic classification unit 122, and a classification learning unit 123. The classification processing unit 121 is a part that performs calculations and determinations regarding sentence classification. The semi-automatic classification unit 122 is a part that performs semi-automatic classification by inputting a keyword for classification and performing classification. The classification learning unit 123 is a part that stores information associating sentences and classification tags as learning data.

  The sentence summarizing unit 140 includes a summarizing processing unit 141 and a summarizing learning unit 142. The summary processing unit 141 is a part that creates a summary from sentences. The summary learning unit 142 is a part that stores information that associates a sentence with a summary sentence.

  The hardware configuration of the text processing apparatus is realized by a general information processing apparatus such as a personal computer shown in FIG.

  The text processing apparatus 10 has a configuration in which a CPU (Central Processing Unit) 202, a main storage device 204, a network I / F 206, a display I / F 208, an input / output I / F 210, and an auxiliary storage I / F 212 are connected by a bus. ing. The text processing apparatus 10 is connected to the text server 400 via the network 60. The sentence server 400 is a server that stores sentences and distributes the sentences to the sentence processing apparatus 10.

  The CPU 202 controls each unit of the text processing apparatus 10 and loads and executes a necessary program in the main storage device 204. The main storage device 204 is usually composed of a volatile memory such as a RAM, and stores a program executed by the CPU 202 and data to be referred to. The network I / F 206 is an interface for connecting to the network 60. The display I / F 208 is an interface for connecting a display device 220 such as an LCD (Liquid Crystal Display). The input / output I / F 210 is an interface for connecting an input / output device. In the example of FIG. 2, a keyboard 230 and a pointing device mouse 232 are connected as input / output devices.

  The auxiliary storage I / F 212 is an interface for connecting an auxiliary storage device such as an HDD (Hard Disk Drive) 250 or an SSD (Solid State Drive).

  The HDD 250 has a large storage capacity, and stores a program for executing this embodiment. The sentence processing apparatus 10 is installed with a sentence analysis program 310, a sentence classification program 320, a classification change program 330, a sentence summary program 340, a summary change program 350, and a report creation program 360.

  The sentence analysis program 310, the sentence classification program 320, the classification change program 330, the sentence summary program 340, the summary change program 350, and the report creation program 360 are a sentence analysis unit 110, a sentence classification unit 120, a classification change unit 130, and a sentence summary, respectively. This is a program for realizing the functions of the unit 140, the summary change unit 150, and the report creation unit 160.

  The HDD 250 stores classification result data 510, summary result data 520, classified / summarized data 530, and sentence analysis result data 540. The significance and structure of each data will be described in detail later.

Next, the outline of the feature of the text processing of the present invention will be described with reference to FIG.
Conventionally, when classifying a sentence, the likelihood / confidence level that applies to the classification is calculated from the sentence based on past classification record data, and the classification is performed.

  In the present invention, a specified classification is input by a user together with a sentence, and semi-automatic classification is performed (phase I in FIG. 3). Here, the designated classification refers to a classification tag name and a keyword for classification newly input by the user. Also, such a method of inputting a keyword for classification and causing the information processing apparatus to classify is called semi-automatic classification. As a result, the classification result that the user wants to extract can be extracted semi-automatically.

  In the past, summary results were generated from text based on parameters. In addition, parameters were updated based on the evaluation results. Here, the parameter is, for example, an amount expressing a feature of a sentence or a set of sentences as a weight as described in Patent Document 2.

  In the present invention, when summarizing a sentence, the structure / meaning of the sentence is analyzed, and information on the classification in phase I in FIG. 3 is added to the information to create a summary sentence (phase II in FIG. 3). As a result, a summary sentence suitable for the classification can be created.

Next, a data structure used in the text processing apparatus according to the present embodiment will be described with reference to FIGS.
The classification result data 510 is data used as a standard for classifying sentences, and has a data format including classification tags 511 and sentences 512 as shown in FIG. This indicates that the tag indicated by the classification tag 511 is classified with respect to the text pattern stored in the sentence 512. This classification result data 510 is generated by learning through classification learning processing (described later).

  The summary performance data 520 is data used as a standard for creating a summary sentence of a sentence. As shown in FIG. 5, a data format including a classification tag 521, a sentence 522, a sentence structure 523, and a summary sentence 524 column. Have This indicates that the summary sentence described in the summary sentence 524 is generated for the text pattern of the sentence 522 classified by the classification tag 521 and having the sentence structure described in the sentence structure 523. The classification result data 510 is generated by learning through summary learning processing (described later).

  The classified / summarized data 530 is data that is stored as a result of classifying sentences and creating a summary by the sentence processing apparatus of the present embodiment. As shown in FIG. 532, a classification tag 1 (533), a classification tag 2 (534), and a summary data 535 column.

  Tags written in classification tag 1 (533) and classification tag 2 (534) are assigned as classification tags to the text stored in text 532 on the registration date described in registration date 531. This shows that the summary sentence described in the summary sentence 535 has been generated. In the description of the present embodiment, two classification tags are used. However, one classification tag may be used, or three or more classification tags may be included.

  The sentence analysis result data 540 is data that holds the analysis result of the sentence included in the sentence to be processed. As shown in FIG. 7, the simple sentence 541, the word 542, the part of speech 543, the sentence structure 544, the meaning It has a data format consisting of 545 fields.

  This is because the sentence described in the simple sentence 541 has the sentence structure described in the sentence structure 544, which is decomposed into the part of speech described in each part of the word 542 and described in the part of speech 543. It has the meaning described in meaning 545 from the structure. For example, in the example of FIG. 7, a single sentence “I want you to broadcast immediately” from a sentence structure “I want you to do” indicates that it has a meaning of (customer) request.

Next, the user interface of the text processing apparatus according to this embodiment will be described with reference to FIGS.
The report creation start screen 700 is a screen that is first displayed for generating and displaying a report of sentence classification and summary processing by the sentence processing apparatus 10, and a classification tag name for classification in a specified period ( Hereinafter, it is a screen for simply inputting a classification keyword. As shown in FIG. 8, the report creation start screen 700 includes display elements of a text target range column 701, a designated classification input column 702, and a report creation start button 705.

  The text target range column 701 is a column for inputting the period of the text to be handled. For example, the registration date and time of the sentence can be the time when the sentence is created or when the sentence is stored in the sentence server 400. The designated classification input field 702 is a field for inputting information for designated classification, and includes a classification tag name input field 703 and a classification keyword input field 704. The classification tag name input field 703 is a field for inputting a classification tag name used for classification. The classification keyword input column 704 is a column for inputting a keyword used for classification. In the example shown in FIG. 8, “device failure” is input as the classification tag name, and “signal” and “failure” are input as the classification keywords. The report creation start button 705 is a button that is selected when a specified classification is input and a report is created and displayed.

  The sentence classification / summary result screen 710 is a screen for displaying the result of sentence classification / summary processing. As shown in FIG. 9, a classification tag display field 711, a number display field 712, and a detailed list display button 713 are set. Has a repeated display element.

  The classification tag display field 711 is a field for displaying the tag name to which the designated classification is input. The number display column 712 is a column that displays the number of sentences belonging to the specified classification. The detailed list display button 713 is a body selected to display a sentence classification / summary detailed list screen 720 (described later) belonging to the designated classification.

  The sentence classification / summary detail list screen 720 is a screen for displaying detailed information of the sentences subjected to the sentence classification / summarization processing and designating the change. As shown in FIG. 10, as shown in FIG. The display includes a summary sentence display field 722, a classification change button 723, and a summary change button 724.

  The original text display field 721 is a field for displaying the original text. The summary sentence display field 722 is a field for displaying a summary sentence of the original sentence. The classification change button 723 and the summary change button 724 are buttons that are selected to display a classification change screen 730 (described later) and a summary change screen 740 (described later), respectively.

  The classification change screen 730 is a screen for changing the designated classification of the text. As shown in FIG. 11, the original text display field 731, the summary text display field 732, the classification designation combo box 733, and the change reflection button 734 are displayed. Has an element.

  The original sentence display column 731 is a column for displaying the original sentence of the sentence. The summary sentence display column 732 is a column for displaying a summary sentence of the sentence. The classification designation combo box 733 is an input display element for changing the classification tag of the sentence. The change reflection button 734 is a button that is selected to give the selected classification tag to the designated sentence.

  In this embodiment, up to three types of classification tags can be specified. However, one or two types of classification tags can be specified, or four or more types can be specified. It may be.

  The summary change screen 740 is a screen for editing and changing the sentence summary sentence. As shown in FIG. 12, the original sentence display field 741, the classification display field 742, the summary sentence editing area 743, and the change reflection button 744. Display elements.

  The original sentence display column 741 is a column for displaying the original sentence of the sentence. The classification display column 742 is a column for displaying the classification tag of the sentence. The summary sentence editing area 743 is a column for displaying a summary sentence of the sentence. The summary sentence editing area 743 displays the contents of the current summary sentence, and the user can change the summary sentence using an input device such as a keyboard. The change reflection button 744 is a display element that is selected to reflect the summary sentence edited in the summary sentence edit area 743.

  The report display screen 750 is a screen for displaying a report created after the designated classification is input. As shown in FIG. 13, a text target range display field 751, a designated classification display field 752, a summary sentence display field 753, a print button. 754 has a display element of a close button 755.

  The text target range display field 751 is a field for displaying the period of the target text input from the report creation start screen 700. The designated category display column 752 is a column for displaying the category tag name and keyword of the designated category input from the report creation start screen 700. The summary sentence display column 753 is a column for displaying the extracted summary sentence. The print button 754 is a button that is selected when printing information displayed on the report display screen 750. The close button 755 is a button that is selected when the report display screen 750 is closed.

Next, processing performed by the text processing apparatus will be described with reference to FIGS.
The sentence information accumulation process shown in the flowchart of FIG. 14 is a process for accumulating sentences and the classification / summary results of the sentences in the sentence processing apparatus.

In the sentence information accumulation process, first, a sentence analysis process is performed (S101). This sentence analysis process will be described in detail later.
Next, sentence classification processing is performed (S102). This sentence classification process will be described in detail later.
Next, sentence summarization processing is performed (S103). This sentence summarization process will be described in detail later.
Next, the result of the sentence classification / sentence summary of S102 and S103 is accumulated in the classified / summarized data 530 shown in FIG. 6 (S104).

Next, the sentence analysis process will be described with reference to FIG.
The sentence analysis process shown in FIG. 15 is a process of analyzing a sentence at the word level / structure level and storing the result, and corresponds to S101 in FIG.
In the sentence analysis process, the sentence processing apparatus 10 receives a sentence from the sentence server 400 via the network 60 (S201).

Next, the word analysis unit 111 of the sentence analysis unit 110 shown in FIG. 1 analyzes the received sentence and stores it in the sentence analysis result data 540 shown in FIG. 7 (S202).
Next, the structure analysis unit 112 analyzes the related structure of the divided words, extracts the meaning thereof, and stores it in the sentence analysis result data 540 (S203).

The sentence classification process shown in FIG. 16 is a process for adding a classification tag for classification to a sentence, and corresponds to the process of S102 in FIG.
In the sentence classification process, first, information on the appearance words and sentence structure stored in the sentence analysis result data 540 shown in FIG. 7 is input to the sentence classification unit 120 (S301).
Next, the processing of S303 to S305 is performed for each classification tag (not shown) held by the text processing device (S302 to S306).

  The classification processing unit 121 calculates the similarity between the sentence and each classification tag using the appearance word (S303). The similarity between the sentence and each classification tag can be calculated, for example, by taking the sum of the number of times that the keyword of the classification tag appears in the sentence. For example, in the classification tag “device failure”, the word group {device failure, device failure, device output} is placed as a keyword, and the total number of sentences is counted by counting the number of occurrences of that word group in the sentence. The similarity divided by the number of words.

  When the similarity is equal to or higher than the predetermined threshold (S304: YES), the process goes to S305, and when the similarity is lower than the predetermined threshold (S304: NO), the evaluation is performed with the next classification tag (S302). .

  When the similarity is equal to or higher than a predetermined threshold (S304: YES), a target classification tag is assigned to the sentence (S305).

The sentence summarization process shown in FIG. 17 is a process for creating a summary sentence from a target sentence, and corresponds to the process of S103 in FIG.
In the sentence summarization process, first, the appearance word and sentence structure information and the classification result (classification tag added in the sentence classification process) stored in the sentence analysis result data 540 shown in FIG. Input (S401).

  Next, the summary processing unit 141 acquires a classification tag added to the target sentence (S402), and repeats the processes of S404 to S410 for all the classification tags (S403 to S411).

  First, the summary processing unit 141 extracts a meaningful single sentence from the sentence analysis result data 540 (S404). A meaningful simple sentence is a simple sentence stored in a simple sentence 541 corresponding to a record in which the meaning 545 field of the sentence analysis result data 540 is not blank. For example, in the example shown in FIG. 7, a single sentence “I want you to broadcast immediately” is stored in the single sentence 541 of the record stored in the “request” of the meaning 545.

Next, the summary processing unit 141 acquires summary performance data 520 having a target classification tag (S405).
Next, S407 to S409 are repeated for each summary performance data 520 having the acquired target classification tag (S406 to S410).

  First, the summary processing unit 141 calculates the similarity between the single sentence of the target sentence and the sentence of the summary performance data 520 (S407). The similarity between the single sentence of the target sentence and the sentence of the summary achievement data 520 is, for example, the number of times that the single sentence word of the target sentence appears in the word of the summary achievement data 520, and the total number of words in the summary achievement data 520 The value divided by can be evaluated as the similarity.

  When the similarity is equal to or higher than the threshold (S408: YES), the process goes to S409. When the similarity is lower than the threshold (S408: NO), the process goes to the next summary performance data 520.

When the similarity is equal to or higher than the threshold (S408: YES), the summary sentence 524 of the summary performance data 520 is calculated as the summary sentence of the target sentence (S409).
Here, a plurality of summary sentences may be extracted from the target sentence, or a summary sentence having a high similarity in S407 may be used as the summary sentence of the target sentence.

The classification learning process shown in FIG. 18 is a process of adding new information to the classification result data 510, which is learning data for associating sentences with classification tags.
In the classification learning process, first, a new classification tag for the target sentence is designated from the classification change screen 730 shown in FIG. 11 (S501).
Next, information on the target sentence and the new classification tag is input to the classification changing unit 130 (S502).
Next, the information of the target sentence and the new classification tag is accumulated in the classification result data 510 (S503).

The summary learning process shown in FIG. 19 is a process of adding new information to summary result data 520 that is learning data for associating a sentence with a summary sentence.
In the summary learning process, first, a summary sentence for the target sentence is described from the summary change screen 740 shown in FIG. 12 (S601).
Next, information on the target sentence and the new summary sentence is input to the summary changing unit 150 (S602).
Next, information on the target sentence and the new summary sentence is accumulated in the summary performance data 520 (S603).

The report creation process shown in FIG. 20 is a process for creating and displaying a report after inputting a specified classification.
In the report creation process, first, the target sentence range (period), classification tag name, and classification keyword are entered from the report creation start screen 700 shown in FIG. 8 (S701).
Next, the report creation unit 160 acquires data having a registration date within the target sentence range (period) input from the classified / summarized data 530 (S702).

Next, the report creation unit 160 aggregates the classified / summarized data 530 with the information of the classification column, and calculates the number of cases (S703).
Next, the process of S705 is repeated for each input designated classification tag (S704 to S706).

In the loop, semi-automatic classification / summarization processing is performed (S705). Details of the semi-automatic classification / summarization process will be described later.
Finally, the report creation unit 160 displays the result on the report display screen 750 shown in FIG. 13 (S707).

The semi-automatic classification / summarization processing shown in FIG. 21 performs classification and summarization processing based on the designated classification after inputting the designated classification, and corresponds to S705 in FIG.
In the semi-automatic classification / summarization process, first, the report creation unit 160 searches the classified / summarized data 530 having the keywords described in the designated classification information (S801).
Next, the classification learning unit 123 accumulates the searched classified / summarized data 530 in the classification result data 510 with the classification tag name of the designated classification tag name (S802).

Next, the classified / summarized data 530 having the registration date of the text target range (period) is input to the classification processing unit 121 (S803).
Then, the processing of S805 to S808 is repeated for each classified / summarized data 530 having the registration date of the text target range (period) (S804 to S809).

  First, the classification processing unit 121 calculates the degree of fit between the target sentence and the designated classification tag (S805). The degree of fit between the target sentence and the designated classification tag is an index indicating the validity when the designated classification tag is the classification tag of the sentence. For example, a keyword associated with the tag name of the designated classification tag is defined, The number of times that the keyword appears in the target sentence can be defined by a value obtained by dividing by the total number of words in the target sentence.

  When the degree of fit between the target sentence data and the designated classification tag is equal to or greater than the prescribed threshold (S805: YES), the process goes to S807, and when the degree of fit between the target sentence data and the designated classification tag is less than the prescribed threshold (S805: NO) ), Go to S804, and process the next classified / summarized data 530.

When the degree of fit between the target sentence data and the specified classification tag is equal to or greater than a predetermined threshold (S805: YES), the classification processing unit 121 assigns the classification tag of the specified classification to the sentence of the classified / summarized data 530. (S807).
Next, the summary processing unit 141 acquires a summary sentence of the corresponding classified / summarized data 530 (S808).

  As described above, in the sentence processing apparatus according to the present embodiment, by inputting the specified classification for the classified and summarized data, by reclassifying, the sentence of the up-to-date topic is selected, In addition, since the summarization process based on the sentence structure is performed, it is possible to obtain a report of a summary sentence that is valuable to the business entity, such as reflecting the user's request.

DESCRIPTION OF SYMBOLS 10 ... Text processing apparatus 110 ... Text analysis part 111 ... Word analysis part 112 ... Structure analysis part 113 ... Text analysis result processing part 120 ... Text classification part 121 ... Classification processing part 122 ... Semi-automatic classification part 123 ... Classification learning part 130 ... Classification Changing section 140 ... sentence summarizing section 141 ... summary processing section 142 ... summary learning section 150 ... summary changing section 160 ... report creation section

Claims (6)

A sentence processing device that classifies sentences by adding a classification tag to the sentences,
Enter the text and the specified classification tag,
For the designated classification tag and the sentence, calculate a degree of fit that is an index indicating the validity when the designated classification tag is the classification tag of the sentence,
A sentence processing apparatus, wherein the designated classification tag is assigned to the sentence when the degree of fit of the sentence with the inputted designated classification tag is equal to or greater than a predetermined threshold. The designated classification tag includes a classification tag name and a keyword,
The sentence processing apparatus according to claim 1, wherein the inputted sentence is a sentence having the keyword of the designated classification tag.   The sentence processing apparatus according to claim 1, wherein the inputted sentence is data to which a classification tag is already added. A sentence processing apparatus that summarizes a sentence and outputs a summary sentence of the sentence,
Holds summary performance data that is a pair of sentences and summary sentences with classification tags,
Perform structural analysis on the single sentence of the input sentence, extract the meaning related to structural analysis,
For each classification tag of the summary record data, a meaningful single sentence related to the structural analysis is extracted, and when the similarity between the input sentence and the summary record data is equal to or higher than a predetermined threshold The sentence processing apparatus characterized in that the summary sentence of the summary performance data is the summary sentence of the input sentence. A sentence processing method for classifying sentences by adding classification tags to sentences,
Enter the text and the specified classification tag,
For the designated classification tag and the sentence, calculate a degree of fit that is an index indicating the validity when the designated classification tag is the classification tag of the sentence,
A sentence processing method comprising: assigning the designated classification tag to the sentence when the degree of fit of the sentence with the inputted designated classification tag is equal to or more than a predetermined threshold. A sentence processing method for summarizing a sentence and outputting a summary sentence of the sentence,
Holds summary performance data that is a pair of sentences and summary sentences with classification tags,
Perform structural analysis on the single sentence of the input sentence, extract the meaning related to structural analysis,
For each classification tag of the summary record data, a meaningful single sentence related to the structural analysis is extracted, and when the similarity between the input sentence and the summary record data is equal to or higher than a predetermined threshold The sentence processing method, wherein the summary sentence of the summary performance data is used as a summary sentence of the inputted sentence.

Images