JP2011053851A - Classification processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily and appropriately classify and display a plurality of sentences. <P>SOLUTION: This classification processor is provided with: a classification method storage means 46 for storing a classification method including conditions for classifying a plurality of utterances and classification item names attached to the utterances classified under the considerations; a voice data reception means 48 and a voice recognition means 50 for acquiring the plurality of utterances as character data; a classification method retrieval acceptance means 41 for specifying a classification method keyword for determining a classification method to be used for classification; a classification method management means 45 for retrieving the classification method from a classification method storage means 46 based on the classification method keyword; an utterance data classification method 47 for classifying the acquired utterances based on the retrieved classification method; and an utterance data management means 51 and an utterance data display control means 49 for displaying the result of classification by the utterance data classification means 47. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a classification processing apparatus that classifies and displays a plurality of sentences according to their contents.

  In a conference where a plurality of users participate, many statements and information are exchanged. In a meeting, for example, a conclusion about a certain problem is given, or important information about planning of a new product is exchanged. Therefore, it is important to record the comments made at the meeting without omission.

  Devices that record important utterances in a conference in time series are known, but simply recording them in time series is inefficient in finding the necessary utterances. On the other hand, a technique is known that records the structure of the speech and the attribute information of the speech together with the speech, and can efficiently search for the speech suitable for the intention of the searcher of the speech ( For example, see Patent Document 1).

Japanese Patent No. 3879786

  However, although the technique described in Patent Literature 1 can search for a statement intended by the searcher or a similar statement from the statement of the conference, what type of statement was made in the conference as a whole I can't figure it out.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique capable of easily and appropriately classifying a plurality of sentences included in a plurality of statements in a meeting, for example. is there.

  To achieve the above object, a classification processing apparatus according to a first aspect of the present invention is a classification processing apparatus that displays a plurality of sentences classified according to their contents, and a predetermined classification unit in the plurality of sentences. A classification method storage means for storing a classification method including a condition for classifying the classification unit, a classification name assigned to the classification unit classified according to the condition, and a sentence acquisition means for acquiring a plurality of sentences as character data A keyword specifying means for specifying a classification method keyword for determining a classification method to be used for classification, a classification method searching means for searching a classification method from a classification method storage means based on the classification method keyword, and an acquired sentence And classifying means for classifying based on the searched classification method, and classification result display means for displaying the classification result by the classification means. According to such a classification processing device, a classification method is searched by a classification method keyword, and the acquired sentence is classified based on the searched classification method, and the classification result is displayed. For this reason, it is possible to easily and appropriately classify a plurality of sentences and display the classification result.

  In the classification processing apparatus, the sentence acquisition means includes voice data input means for inputting voice data of a sentence issued by a user, and character data conversion means for converting the voice data into character data by voice recognition. Also good. According to such a classification processing device, it is possible to appropriately classify sentences issued by the user.

  In the classification processing apparatus, the voice data input means may be capable of inputting voice data of sentences issued from a plurality of users. According to such a classification processing apparatus, sentences issued from a plurality of users can be collected and classified appropriately.

  In the classification processing apparatus, the classification result display means may display the classification result in a form that can be confirmed by a plurality of users. According to such a classification processing apparatus, a plurality of users can confirm the classification result.

  In the classification processing apparatus, a classification method modification accepting unit that accepts designation of a classification unit for changing the classification and designation of a change destination of the classification unit, and changes the classification result based on the designation. Based on the change display control means to be displayed and the classification method used for the classification, a new classification method having the generated classification condition is generated by generating a classification condition in which the classification unit designated to be changed is classified as the change destination. You may make it have a classification method production | generation means memorize | stored in a classification method memory | storage means. According to the classification processing apparatus, if a classification unit and a change destination for changing the classification are specified for the classification result, a new classification method capable of performing classification according to the change can be generated. This classification method can be used at the time of classification.

  The classification processing apparatus may further include a keyword extraction unit that extracts a keyword from the acquired sentence, and a classification method keyword determination unit that determines a classification method keyword based on the keyword. According to such a classification processing apparatus, it is possible to search for a classification method for performing appropriate classification based on the keywords of the sentence to be classified.

It is a block diagram of the conference system which concerns on one Embodiment of this invention. It is a functional block diagram of the conference system which concerns on one Embodiment of this invention. It is a figure explaining the classification method management information which concerns on one Embodiment of this invention. It is a figure explaining an example of the classification method concerning one embodiment of the present invention. It is a figure explaining the problem management information which concerns on one Embodiment of this invention. It is a figure explaining statement management information concerning one embodiment of the present invention. It is a flowchart of the process at the time of speech which concerns on one Embodiment of this invention. It is a flowchart of the speech classification process which concerns on one Embodiment of this invention. It is a figure explaining an example of the screen at the time of speech which concerns on one Embodiment of this invention. It is a figure explaining an example of the classification result screen concerning one embodiment of the present invention. It is a figure explaining classification change of a statement concerning one embodiment of the present invention. It is a figure explaining the production | generation and addition of the new classification method which concerns on one Embodiment of this invention. It is a figure explaining an example of the classification result screen concerning the modification of the present invention.

  Embodiments of the present invention will be described with reference to the drawings. The embodiments described below do not limit the invention according to the claims, and all the elements and combinations described in the embodiments are essential for the solution of the invention. Is not limited.

  First, a conference system according to an embodiment of the present invention will be described.

  FIG. 1 is a configuration diagram of a conference system according to an embodiment of the present invention.

  The conference system 1 includes an application server 2 as an example of a classification processing apparatus and a user terminal 3 that is used by a plurality of users, connected via a network 4. The application server 2 is configured by, for example, a PC (Personal Computer). The user terminal 3 is configured by a PC, for example. The network 4 may be a LAN (Local Area Network), a WAN (Wide Area Network), or may include the Internet. In the present embodiment, the network 5 includes the Internet, for example, and can access various servers (not shown) connected to the Internet.

  FIG. 2 is a functional configuration diagram of the conference system according to the embodiment of the present invention.

  The user terminal 3 includes an input unit 31 and a display unit 32.

  The input unit 31 includes, for example, a keyboard, a mouse, a microphone, and the like, and inputs various instructions by the user and voices of statements (sentences) uttered by the user. The display unit 32 includes, for example, a display device, and causes the display device to display an image corresponding to the data based on data from the application server 2.

  The application server 2 includes a classification method search reception unit 41, a classification method determination reception unit 42, a classification method correction reception unit 43, a classification method display control unit 44, a classification method search unit, and a classification as an example of a keyword specifying unit. Classification method management means 45 as an example of method generation means, classification method storage means 46, speech data classification means 47 as an example of classification means, voice data reception means 48 as an example of voice data input means, and classification Speech data display control means 49 as an example of result display means and change display control means, speech recognition means 50 as an example of character data conversion means, speech data management means 51, keyword extraction means 52, speech data storage Means 53.

  The classification method search reception means 41 receives a keyword (classification method keyword) for searching a classification method used for classification input by the user from the input means 31 and passes it to the classification method management means 45. The classification method determination accepting means 42 accepts a classification method designated from a plurality of search methods by the user from the input means 31 and passes it to the classification method management means 45. The classification method correction accepting unit 43 accepts a statement designated by the user to change the classification and a set of classifications to be changed. The classification method display control unit 44 controls the display unit 32 of the user terminal 3 to display the search result classification method and the like based on the notification from the classification method management unit 45 (transmission of information necessary for display). I do.

  The classification method management means 45 searches the classification method storage means 46 based on the classification method keyword received from the classification method search accepting means 41, detects a classification method that is a use candidate, and sends it to the classification method display control means 44. hand over. Further, the classification method management unit 45 acquires information on the classification method from the classification method storage unit 46 based on the designation of the classification method to be used received from the classification method determination reception unit 42 and notifies the comment data classification unit 47 of the information. To do. The classification method management means 45 receives from the classification method correction accepting means 43 the designation of correction between the statement for changing the classification and the set of classifications to be changed, and instructs the comment data classification means 47 for the correction. A classification method corresponding to the correction is generated and stored in the classification method storage unit 46.

  The classification method storage means 46 stores classification method management information.

  FIG. 3 is a diagram illustrating classification method management information according to an embodiment of the present invention.

  As shown in FIG. 3, the classification method management information stores a classification method ID, a classification method name, and a classification method. The classification method ID is identification information for identifying the classification method. The classification method name is the name of the classification method. As the classification method, in this embodiment, a classification set name indicating each set to be classified and a condition (classification condition) classified into the set are directly stored, and a combination of other classification methods is used. There is.

  4A and 4B are diagrams for explaining an example of a classification method according to an embodiment of the present invention. FIG. 4A shows a classification method with classification method ID = 2, and FIG. 4B shows a classification method with classification method ID = 3. 4C shows a classification method with classification method ID = 4, and FIG. 4D shows a classification method with classification method ID = 5.

  The classification methods with classification methods ID = 2, 3, and 4 include a classification set name and classification conditions. For example, as shown in FIG. 4B, the classification method with classification ID = 3 is classified as “serious” when “severe”, “serious”, “urgent”, or “pretty” is included in the utterance. This means that if “minor” and “problem” are included, it is classified into the set of the category set name “minor”. The classification method with classification method ID = 5 means that the classification methods with classification methods ID = 3 and 4 are combined, and specifically, “serious”, “difficult to handle”, “serious” and This indicates that the classification is made into four groups of “easy to deal”, “minor” and “difficult to deal”, “minor” and “easy to deal”.

  The utterance data classification unit 47 performs processing for classifying the utterance data stored in the utterance data storage unit 53 in accordance with the classification method notified from the classification method management unit 45. In the present embodiment, the utterance data classification means 47 uses a single utterance of a single user, that is, a word, phrase, sentence, or sentence that is continuously uttered by the user once as a classification unit. The voice data receiving means 48 receives speech voice data input by the input means 31 of the user terminal 3. The voice recognition means 50 recognizes the voice of the voice data received by the voice data reception means 48, generates text-format speech data, and passes it to the speech data management means 51. The utterance data management unit 51 stores the utterance data received from the voice recognition unit 50 in the utterance data storage unit 53. Further, the utterance data management means 51 acquires the utterance data stored in the utterance data storage means 53 and its classification information, and passes them to the utterance data display control means 49. Based on the utterance data received from the utterance data display control means 49, the utterance data management means 51, and the classification information, the classification result of the utterance data is displayed on the display means 32 of the user terminal 3.

  The keyword extraction means 52 extracts keywords from the utterance data stored in the utterance data storage means 53 and stores them in the utterance data storage means 53 in association with the utterance data. Here, as a method of extracting a keyword from the utterance data, for example, the utterance data may be analyzed by morphological analysis, and a predetermined part of speech may be used as a keyword, and words, phrases, and the like that become keywords are stored in a dictionary. When a word that matches the stored word, phrase, or the like is included, it may be used as a keyword of the utterance data, or these may be combined.

  The utterance data storage unit 53 stores task management information and utterance management information.

  FIG. 5 is a diagram illustrating problem management information according to an embodiment of the present invention, and FIG. 6 is a diagram illustrating statement management information according to an embodiment of the present invention.

  The task management information includes a task ID for identifying the task and a title that is a summary of the speech that is the task. The speech management information includes a speech ID for identifying a speech, a task ID indicating a task related to the speech, speech content (speech data), a keyword included in the speech, a classification method, and speech data corresponding to the classification. Classification set name classified by the method. A plurality of classification methods and classification set names can be stored, and when a new classification method is used for a statement, the classification method and the classified classification set name are registered. It has become. When there is no set name to be classified, the classified set name is blank.

  Next, the operation in the conference system 1 will be described.

  FIG. 7 is a flowchart of the speech processing according to an embodiment of the present invention.

  The utterance data management means 51 of the application server 2 determines whether or not the utterance of the task has been received from the user of the conference host (step S1). Here, whether or not the assignment has been accepted is, for example, whether a specific term indicating the assignment is included in the speech data received by the voice data receiving unit 4 from the user terminal 2 of the moderator Alternatively, the determination may be made according to whether or not the moderator user is performing a predetermined operation with a mouse or the like.

  As a result, when the task is not accepted, the process waits, while when the task is accepted (step S1: YES), the speech data management means 51 is received by the voice data receiving means 48, and the voice recognition means. The title is extracted from the utterance data recognized by 50 and registered in the problem management information of the utterance data storage means 53 (step S2), and the utterance data is further registered in the utterance management information (step S3). The keywords included in the utterance data are extracted and registered in the utterance management information (step S4). At this time, the utterance data management means 51 causes the utterance data display control means 49 to additionally display the title 91 and its utterance 92 on the screen 90 of the display means 32 of the user terminal 3, as shown in FIG. .

  Thereafter, the utterance data management means 51 determines whether or not there is a utterance based on whether or not the utterance data based on the voice data from the user terminal 3 of the participant has been received (step S5). Is registered in the speech management information (step S6), and the keywords included in the speech data are extracted and registered in the speech management information (step S7). At this time, the utterance data management means 51 causes the utterance data display control means 49 to additionally display the utterance 92 corresponding to the display means 32 of the user terminal 3 as shown in FIG.

  Next, the utterance data management means 51 determines whether or not the end of the utterance regarding the current task has been received from the user of the conference presenter (step S8). Here, whether or not the end of the speech about the task has been accepted is, for example, a specification indicating the end of the speech about the task in the speech data of the speech received by the speech data receiving means 48 from the user terminal 3 of the moderator May be determined based on whether or not the term is included, or may be determined based on whether or not a predetermined operation is performed with a mouse or the like.

  As a result (step S8), when the end of the speech about the task is not received, the processing from step S5 is continuously executed, and the processing for receiving the speech for the same task is performed.

  On the other hand, when the end of the speech about the task is accepted (step S8: YES), the speech data management means 51 determines whether or not to proceed to another task (step S9). Here, as to whether or not to proceed to another task, for example, a specific term indicating that the task proceeds to another task in the speech data of the speech received by the speech data receiving unit 48 from the user terminal 3 of the presenter May be determined depending on whether or not a predetermined operation is performed by using a mouse or the like. As a result, when it is determined that the task proceeds to another task (step S9: YES), the process proceeds to the process from step S1 and proceeds to the process of accepting a speech for the next task, but is determined not to proceed to another task. If so, the process ends. Through this processing, the task management information and the speech management information are accumulated in the speech data storage means 53, and thereafter, the speech classification processing described below is performed.

  FIG. 8 is a flowchart of speech classification processing according to an embodiment of the present invention.

  First, the classification method search reception unit 41 receives a classification method keyword input by the user from the input unit 31 of the user terminal 3 (step S11), and passes the classification method keyword to the classification method management unit 45. Thereby, the classification method management means 45 specifies the classification method keyword (step S11).

  The classification method management means 45 searches the classification method storage means 46 for a classification method corresponding to the classification method keyword, and causes the classification method display control means 44 to display candidate classification methods on the display means 32 of the user terminal 3. (Step S12). The classification method management unit 45 may search for a classification method from a server (not shown) connected to the network 4. Further, as a classification method search based on a classification method keyword, a classification method keyword or a keyword related to the classification method keyword may be searched, and a classification method may be used as a classification method condition. You may make it search the classification method by which the keyword or the keyword relevant to this is set.

  Next, the classification method management unit 45 determines whether or not the selection of the classification method has been received from the user terminal 3 (that is, the user) via the classification method determination reception unit 42 (step S13). As a result, if the selection is not accepted, the process is awaited.

  On the other hand, when the selection is accepted (step S13: YES), the classification method management unit 45 takes out the selected classification method from the classification method storage unit 46 and passes it to the speech data classification unit 47. The utterance data classification means 47 classifies each utterance in the utterance data storage means 53 based on the received classification method (step S14), reflects it in the utterance management information, and indicates that the categorization has been performed. The management unit 51 is notified. The utterance data management means 51 generates an image indicating the classification result of categorizing the utterances based on the utterance management information in the utterance data storage means 53, and the utterance data display control means 49 displays the image on the display means 32 of the user terminal 3. It is displayed (step S15). Therefore, the speech can be easily classified and displayed by the classification method desired by the user.

  FIG. 10 is a diagram illustrating an example of a classification result screen according to an embodiment of the present invention. Here, FIG. 10 is a classification result screen when the classification method of classification method ID = 5 in FIG. 3 is selected in step S13.

  The classification result screen 100 includes an agenda display area 101, a classification method display area 102, a serious / difficult to cope area 103, a serious / easy to cope area 104, a minor / difficult to cope area 105, and a minor / easy to cope area 106. Is displayed. The agenda is displayed in the agenda display area 101. In the classification method display area 102, the name of the classification method used for classification is displayed. In the serious / hard to cope area 103, messages classified into a set of serious and difficult to handle are displayed. In the serious / easy correspondence area 104, messages classified into a serious and easy correspondence group are displayed. In the minor / difficult-to-handle area 105, messages classified into a minor and difficult-to-handle set are displayed. In the minor / easy-to-handle area 106, messages classified into minor and easy-to-handle groups are displayed.

  Returning to the description of FIG. 8, the classification method management unit 45 determines whether the classification method search reception unit 41 has received a change of the classification method from the user terminal 3 (step S <b> 16), and has accepted the change of the classification method. In the case (step S16: YES), the process returns to step S11.

  On the other hand, when the change of the classification method has not been received (step S16: NO), the classification method management unit 45 determines whether or not the classification method correction reception unit 43 has received an instruction to correct the classification method ( Step S17). As an instruction to modify the classification method, there is an instruction to move to another set area for a certain statement. As an example of a method of accepting an instruction to correct the classification method by the input means 31 in the user terminal 3, as shown in FIG. 11, a message 110 for correcting the classification is changed from the area 104 that was originally classified to another area 103. An instruction can be received by a mouse operation to be dragged and dropped.

  As a result, when it is determined that an instruction to correct the classification method has been received (step S17: YES), the utterance data classification unit 47 corrects the classification in the utterance management information of the utterance data storage unit 53 according to the instruction. Then, the utterance data management means 51 causes the utterance data display control means 49 to display the corrected classification result on the display means 32 based on the corrected classification result (step S18). Next, the classification method management unit 45 corrects the classification method in accordance with the correction instruction (step S19), and adds the corrected classification method to the classification method storage unit 46.

  Here, correction of the classification method by the classification method management means 45 will be described.

  The classification method management unit 45 changes the conditions in the classification method so that the remarks to be corrected are classified into a set of destination areas. For example, it may be added as a parallel condition that the keyword in the utterance to be corrected is included in the classification condition classified into the destination set, or it is classified into the source set. A condition for excluding a keyword included in a remark to be corrected may be added to the classification condition. Further, when there are some conditions corresponding to the correction instruction, any one of them may be arbitrarily selected, or the user may be allowed to select which one is the condition.

  For example, when an instruction to correct the classification method is generated as shown in FIG. 11, the classification method management unit 45, as shown in FIG. 12B, the classification method before the change (the classification method ID = 3 in FIG. 4B). Classification method classified as “serious” in the category, with “large” included in the remarks to be corrected as a condition added as a classification method (classification method ID = 6) different from that before the change It is added to the storage means 46. Furthermore, the classification method management unit 45 newly adds a classification method (classification method ID = 6) used in the classification method before change (classification method ID = 5 in FIG. 4D). The classification method changed to () (FIG. 12C: Classification method ID = 7) is newly added to the classification method storage means 46. As a result, the classification method reflecting the correction of the current classification method can be easily used thereafter.

  Returning to the description of FIG. 8, after step 19, the classification method management unit 45 determines whether or not there is an instruction to end the classification (step S20). If there is no instruction to end, the process returns to step S16. On the other hand, if the end instruction is met, the process ends.

  Although the present invention has been described based on the embodiments, the present invention is not limited to the above-described embodiments, and can be applied to various other modes.

  For example, in the above embodiment, classification is made for each utterance, but the present invention is not limited to this, and as shown in FIG. May be classified. Here, in the case of a classification method combining a plurality of classification methods (for example, a classification method combining a classification method for serious / minor and a classification method for easy / hard to cope), either one of them is used. When a classification method (for example, a classification method for serious / minor) is classified and the other classification method cannot be classified (a keyword that includes or contradicts a keyword that contradicts each condition of the classification method) 13), as shown in the task 111 of FIG. 13, the task may be displayed across a plurality of areas regarding the classification method that cannot be classified. In FIG. 13, the task 111 is an example of a task that is classified as minor but cannot be classified as either easy to handle or difficult to handle, and is displayed so as to span the areas 105 and 106 corresponding to the minor. In this case, the assignment of the assignment 111 is accepted by the user's operation, the assignment classification is corrected according to the operation result, and the comment data management means 51 controls the comment data display based on the corrected classification result. The corrected classification result is displayed on the display means 32 by the means 49, and the classification method management means 45 corrects the classification method in accordance with the correction instruction, and adds the corrected classification method to the classification method storage means 46. You may make it do.

  Further, in the above-described embodiment, the screen of the content to be shared such as the classification result is displayed on each of the plurality of user terminals 3, but the present invention is not limited to this, for example, a shared screen such as a projector In an environment where each user can see a screen displayed on the display device, the shared screen display device may display a screen having contents to be shared. Moreover, although the said embodiment demonstrated the example which used the microphone connected to the several user terminal 3 as the input device which inputs each user's audio | voice, for example, the meeting in one conference room is performed. If so, a plurality of users' voices may be input from one microphone.

  Further, in the above-described embodiment, a plurality of users make a statement, and after that, the classification method is determined, the classification is performed, and the classification result is displayed. However, the present invention is not limited to this, for example, in advance A classification method may be determined, and each time a user speaks, the speech is classified, and the classification result of the speech may be added to the screen. In the above embodiment, the classification method is searched based on the classification method keyword specified by the user. However, the present invention is not limited to this. For example, the classification method management means 45 (the classification method keyword determination means) As an example, the classification method keyword may be determined based on the keyword in the utterance data stored in the utterance data storage means 53 to search for the classification method. In this case, for example, one or more keywords are determined as classification method keywords based on the frequency of appearance of keywords in the utterance data, and a classification method having a high relevance to the classification method keyword is searched. May be.

  Moreover, in the said embodiment, when the user's evaluation with respect to the used classification method is collected and the candidate of a classification method is shown, you may make it consider the height of the said evaluation. In addition, when displaying the classification result of an utterance, the classification result by a different classification method may be displayed at an arbitrary timing. In this way, it is possible to give the user a new awareness that the user was not aware of. In the above embodiment, an example in which a speech by a user's voice is input and classified is shown. However, the present invention is not limited to this, and when a sentence (or sentence) is classified with respect to text-format document data. The present invention can also be applied to.

  In the above embodiment, one utterance of each user is classified as a unit. However, the present invention is not limited to this. For example, a plurality of utterances for the content designated by the presenter are used as one unit. In summary, each unit may be classified.

DESCRIPTION OF SYMBOLS 1 Conference system, 2 Application server, 3 User terminal, 4 Network, 31 Input means, 32 Display means, 41 Classification method search reception means, 42 Classification method determination means, 43 Classification method correction reception means, 44 Classification method display control means 45, classification method management means, 46 classification method storage means, 47 speech data classification means, 48 speech data reception means, 49 speech data display control means, 50 speech recognition means, 51 speech data management means, 52 keyword extraction means, 53 speech Data storage means.

Claims (6)

A classification processing device for classifying and displaying a plurality of sentences according to their contents,
A classification method storage means for storing a classification method including a condition for classifying a predetermined classification unit in a plurality of sentences, and a classification name assigned to the classification unit classified according to the condition;
Sentence acquisition means for acquiring a plurality of sentences as character data;
A keyword identification means for identifying a classification method keyword to determine a classification method to be used for classification;
A classification method search means for searching for a classification method from the classification method storage means based on the classification method keyword;
Classifying means for classifying the acquired sentence based on the searched classification method;
A classification processing apparatus having classification result display means for displaying a classification result by the classification means. The sentence acquisition means includes:
Voice data input means for inputting voice data of a sentence issued by a user;
The classification processing apparatus according to claim 1, further comprising: character data conversion means for converting the voice data into character data by voice recognition. The voice data input means includes
The classification processing apparatus according to claim 2, wherein voice data of sentences issued from a plurality of users can be input. The classification result display means includes
The classification processing apparatus according to any one of claims 1 to 3, wherein the classification result is displayed in a form that can be confirmed by a plurality of users. A classification method correction accepting means for accepting designation of a classification unit for changing the classification and designation of a change destination of the classification unit for the classification result;
Change display control means for changing and displaying the classification result based on the designation;
Based on the classification method used for classification, a classification condition is generated in which the classification unit designated to be changed is classified as the change destination, and a new classification method having the generated classification condition is stored in the classification method storage unit. The classification processing apparatus according to claim 1, further comprising: a classification method generation unit that stores the classification method. Keyword extracting means for extracting keywords from the acquired sentence;
The classification processing apparatus according to claim 1, further comprising: a classification method keyword determination unit that determines the classification method keyword based on the keyword.

Images