JP2018045326A - Record dividing device and record dividing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce work for a user to divide a message record in comparison with a case where the user themselves reads the whole message record.SOLUTION: A record dividing system of the present invention includes: a topic change estimation unit for obtaining a message record of a chat; a file operation history storage unit for storing an operation history, i.e. a history of operation of a client by a user; and a minute creation unit for dividing a message record on the basis of common operation which is operation recorded in the operation history, executed at a time period the message of the chat is transmitted and received, and common to at least two or more users.SELECTED DRAWING: Figure 10

Description

  The present invention relates to a recording division apparatus and a recording division system.

  Patent Document 1 discloses a conference management unit that manages minutes, a network control unit that performs control to receive display data from a portable information terminal, and a screen that performs control to display a display screen based on the received display data. Based on the attached information of the received display data, the control unit, the agenda switching detection unit for detecting the switching of the agenda, and the minutes managed by the conference management unit, the switching is detected by the agenda switching detection unit. An electronic conference system including an agenda document creation unit that creates agenda information for each agenda is described.

Japanese Patent Laying-Open No. 2015-232769

  By the way, in recent years, so-called chat has been increasingly used as a communication tool using a telecommunication line. And the content of communication using such a chat etc. may be transmitted to the communication participant or its related person.

  Here, if you want to convey the contents of the communication to only a part of the participants and all concerned parties, in other words, do not want to convey a part of the contents of the communication to some of the parties concerned Cases are also envisaged. In this case, it is necessary to divide the message record indicating the content of communication for each content. However, if the user divides the message record, for example, the user needs to read the entire message record and determine a point to be divided, which places a heavy burden on the user.

  Therefore, an object of the present invention is to reduce the work for dividing the message record by the user as compared with the case where the user himself / herself reads the entire message record and divides the message record.

The invention according to claim 1 is connected to the telecommunication line, an acquisition unit for obtaining a message record in which a plurality of messages transmitted and received via the telecommunication line are arranged in time series among a plurality of users. A storage unit that stores an operation history that is a history of a user's operation of the apparatus; an operation that is recorded in the operation history, and is executed when a plurality of messages in the message record are transmitted and received; and at least two people A recording / dividing apparatus comprising: a dividing unit that divides the message recording based on a common operation that is an operation common to the above users.
The invention according to claim 2 includes another storage unit that stores an information exchange history that is a history of information exchange among a plurality of parties including the plurality of users, and the dividing unit includes the common operation and The recording and dividing apparatus according to claim 1, wherein the message record is divided based on the information exchange history.
The invention according to claim 3 is another storage unit that stores an information exchange history that is a history of information exchange among a plurality of parties including the plurality of users, and the party that is recorded in the information exchange history. And a specifying unit that specifies a word related to each information exchange, and the dividing unit includes at least one of the related person and the related word included in the message record, and the information exchange history. 2. The recording dividing apparatus according to claim 1, wherein the message recording is divided on the basis of the message recording.
The invention according to claim 4 has a set specifying unit for specifying a plurality of sets of the related parties and the related words in the information exchange history, and the dividing unit is the related parties included in the message record. And at least one of the related words, and the set including the specified one is different from the set in the message before the message including the specified related party or the related word 4. The recording and dividing apparatus according to claim 3, wherein the message recording is divided between the previous message and the message including the one.
According to a fifth aspect of the present invention, information of a person who is at least one of the parties concerned and is related to the content of the divided message record is used as a transmission destination candidate for transmitting the divided message record. 5. The display device according to claim 2, further comprising: a display unit for displaying, and a reception unit that receives an instruction to transmit the divided message record to the related person displayed on the display unit. The recording division apparatus according to the item.
The invention according to claim 6, wherein the dividing unit divides the message record based on an operation of the same file by the at least two users as the common operation. 6. The recording dividing apparatus according to any one of 5 above.
The invention according to claim 7 is an acquisition unit that acquires a message record in which a plurality of messages transmitted and received via a telecommunication line are arranged in time series among a plurality of users; A recording division comprising: an extraction unit for extracting a turning point at which a topic referred to in the message has changed; and a dividing unit for dividing the message record based on the turning point extracted by the extraction unit Device.
According to an eighth aspect of the present invention, a providing unit that provides a service for transmitting and receiving a plurality of messages via a telecommunication line among a plurality of users, and a message record in which the plurality of messages are arranged in time series are acquired. An acquisition unit; a storage unit that stores an operation history that is a history of a user operating a device connected to the telecommunication line; and an operation recorded in the operation history, wherein a plurality of messages in the message record are A recording division system comprising: a division unit that divides the message record based on a common operation that is executed at a time of transmission / reception and is common to at least two or more users.

According to the first aspect of the present invention, the user's work for dividing the message record can be reduced as compared with the case where the user himself / herself reads the entire message record.
According to the second aspect of the present invention, message recording can be divided in messages related to parties other than users included in the message recording.
According to the third aspect of the present invention, it is possible to divide a message record in a message including a related person other than the user included in the message record or a related word.
According to the invention of claim 4, message recording can be divided in a message whose topic included in message recording is changed.
According to the invention of claim 5, transmission work by the user of the divided message record is facilitated.
According to the sixth aspect of the present invention, message recording can be divided based on the same file operation performed by a plurality of users.
According to the seventh aspect of the present invention, the user's work for dividing the message record can be reduced as compared with the case where the user himself / herself reads the entire message record.
According to the invention of claim 8, it is possible to reduce the work for dividing the message record by the user as compared with the case where the user himself reads the entire message record.

It is the figure which showed the example of whole structure of the recording division | segmentation system with which embodiment is applied. It is a figure for demonstrating the structural example of a server. An example of message recording is shown. An example of a topic list is shown. It is the flowchart which showed the example of creation operation | movement of the topic list by a server. It is the flowchart which showed topic change presumption operation based on file operation history. It is the flowchart which showed the topic change estimation operation | movement based on the person word contained in a message recording. (A) And (b) is a figure for demonstrating the operation | movement which acquires an applicable record list. (A) And (b) is a figure for demonstrating the operation | movement which acquires a word extraction list. (A) thru | or (d) is a figure for demonstrating the specific example of the topic list creation operation | movement by a server. It is a figure for demonstrating the structural example of the user interface for the minutes preparation. (A) And (b) is a figure for demonstrating the structural example of a user interface. It is the figure which showed the hardware structural example of the server.

Embodiments of the present invention will be described below with reference to the accompanying drawings.
<Recording division system 1>
FIG. 1 is a diagram showing an example of the overall configuration of a recording division system 1 to which the embodiment is applied.
The recording division system 1 includes a client group 10 including a first client 11, a second client 12, and a third client 13, and a server 20. Further, the client group 10 and the server 20 are connected via a network 40. The recording division system 1 is an example of a recording division apparatus.

  Each of the first client 11 to the third client 13 configuring the client group 10 is configured by a computer device such as a desktop computer, a notebook computer, a tablet computer, or a smartphone. In the illustrated example, the client group 10 is composed of three clients, but the number of clients is not limited to this.

The server 20 is configured by a computer device. The server 20 of the present embodiment has a function of processing messages (electronic messages) exchanged between users in a service (user service) for users provided to the first client 11 to the third client 13. Yes.
Here, in the illustrated example, the service for the user is so-called chat. Further, the subject (providing unit) that provides the chat service may be the server 20 or a computer device other than the server 20 included in the network 40.

  Here, chat is a service (service software) for exchanging messages among a plurality of people through a telecommunication line. In addition, chatting allows users to discuss, discuss, or hold a meeting. Note that chat can be understood as a system in which a plurality of users exchange text data while connected to a network. Moreover, the message exchanged by chat can be grasped as an utterance of each user.

  In addition, chat may be provided in community-based membership services such as social network services (SNS). In chat, a plurality of people can exchange short sentences in real time. The exchanged sentences are displayed on one screen (displayed as a list) on the clients (first client 11 to third client 13) used by each user. The message is composed of text data as described above, and does not include information such as character type (font), size, and character modification.

In general, chat has an advantage that messages can be exchanged easily and easily. For this reason, there are increasing opportunities to be used when working from a remote location such as at home or abroad. It may also be used instead of talking.
On the other hand, in order to widely deploy and store information, email is often used instead of chat. In other words, chat and e-mail have different purposes and tool forms, and are used according to their characteristics.

  Here, there is a case where the content (exchanged message group) discussed in the chat is to be transmitted to the chat participant or a person who is not a participant but has the content. In this case, for example, if the entire content discussed in the chat is transmitted as it is by e-mail, there may be a case where there is a security problem. In particular, when there are a plurality of topics in the content discussed in the chat, and related parties are different for each topic, it may be considered that it is not preferable to transmit a specific topic to all the related parties.

  Therefore, in the present embodiment, the content discussed in the chat is divided for each topic, and information is transmitted by transmitting an e-mail only to a person related to the topic. In further explanation, a topic turning point (hereinafter referred to as a turning point) in chat is estimated from the behavior history of chat participants, work relationships between users, human relationships, and the like. Then, the content discussed in the chat at this turning point is divided and transmitted to the parties concerned.

Here, the topic in chat is a subject (agenda, topic) to be discussed or dialogued with messages exchanged in chat.
The turning point is a break when the topic in chatting is different. In the following explanation, the turning point is set for convenience in the first message when moving from the previous topic to the next topic. That is, the first message that moves to the next topic is recognized as a turning point. Note that unlike this example, the last message on the previous topic, or between the last message on the previous topic and the first message on the next topic may be used as a turning point.

  In addition, the turning point in the present embodiment can also be understood as a point in which a person concerned who is referred to in each message changes in a series of message groups (message records). Further, the number of turning points included in the message record varies depending on the content of the message. Thus, there may be no turning points included in the message record, and there may be one or more turning points.

In the present embodiment, the turning point is estimated based on the file operation of the chat participant, and the turning point is estimated based on the person (person name) and the word (keyword) in the contents discussed in the chat. .
In addition, the inference of the turning point based on the file operation is premised on the existence of the material that is the subject of the topic in the chat. In other words, it detects that the topic has changed by inferring a turning point from the operation history of the file.

  In addition, inferring a turning point based on a person / word detects a person / word that may be a topic from the communication history of a chat participant, without presuming the existence of a topic subject in chat. By detecting the turning point, it is detected that the topic has changed. In addition, by performing processing that combines a person and a word, even when a specific word or a specific person appears in different utterances, it can be recognized that they are the same topic (details will be described later). ).

  Note that in communication using a communication tool such as chat, messages are not necessarily exchanged using materials such as documents and tables. To further explain, there may be a case where a topic having a document and a topic having no document are mixed.

  Therefore, for example, if the turning point in the chat recording is specified based only on the material that is the subject of the topic, it is difficult to specify the turning point for the topic that does not use the material. However, in the present embodiment, topic switching can be detected even when there is no material subject to the topic.

<Server 20>
<Configuration Example of Server 20>
FIG. 2 is a diagram for explaining a configuration example of the server 20.
Next, a configuration example of the server 20 will be described with reference to FIGS. 1 and 2.
The server 20 according to the present embodiment includes an input / output unit 200, an operation information acquisition unit 201, a topic list creation unit 203, a minutes creation unit 205, a transmission unit 207, a topic conversion estimation unit 210, a file operation A history promotion unit 221 and a communication history storage unit 223 are included.

The input / output unit 200 inputs / outputs various data to / from the network 40.
The operation information acquisition unit 201 acquires operation information that is information related to the operation of the client (the first client 11 to the third client 13) by the user. This operation information includes file operation information and communication content information.

  The topic list creation unit 203 creates a topic list in which a list of topics (utterance records) made by a plurality of users is described. Although details will be described later, a topic list is created based on the turning points estimated by the topic change estimating unit 210. The topic list creation unit 203 is an example of a specifying unit and a set specifying unit.

The minutes creation unit 205 divides the message record based on the topic list created by the topic list creation unit 203. In other words, the minutes creation unit 205 creates the minutes from the message record based on the topic list.
In addition, in the following explanation, the minutes of the message records are divided. In other words, multiple minutes are created from the message record. As an exception, if the message is not divided, the message record becomes the minutes as it is.
The minutes creation unit 205 is an example of a division unit.

The transmitting unit 207 transmits the minutes created by the minutes creating unit 205 to the user via the input / output unit 200. In other words, the transmission unit 207 assists transmission to the user. The transmission unit 207 displays a UI (user interface, which will be described later) on the first client 11 to the third client 13. Note that the minutes creation unit 205 creates the minutes when the user's operation is accepted by any UI of the first client 11 to the third client 13.
The transmission unit 207 holds template information such as an email transmitted to the user. In addition, the transmission unit 207 creates an email to which the minutes are attached according to the template information. The transmission unit 207 is an example of a display unit and a reception unit.

  The topic conversion estimation unit 210 estimates a conversion point in message recording. Details of the inference of the turning point will be described later. The topic conversion estimation unit 210 includes a file operation estimation unit 211 and a person word estimation unit 213. The topic change estimation unit 210 is an example of an acquisition unit and an extraction unit that acquire message records.

The file operation estimation unit 211 estimates a turning point from file operation information by a chat participant.
The person word estimation unit 213 estimates a turning point from the chat content. Specifically, a turning point is estimated based on a word included in the chat content or a related person (related person name). In the illustrated example, the person word estimation unit 213 will be described as including a word extraction unit 215 that extracts words included in chat content and a related person extraction unit 217 that extracts related persons.

The file operation history storage unit 221 stores information regarding the file operation acquired by the operation information acquisition unit 201. That is, the file operation history storage unit 221 stores a file operation history. The file operation history storage unit 221 is an example of a storage unit.
The communication history storage unit 223 stores information related to communication contents acquired by the operation information acquisition unit 201. That is, the communication history storage unit 223 stores message records and communication histories. The communication history storage unit 223 is an example of another storage unit. The communication history is an example of an information exchange history.

FIG. 3 shows an example of message recording.
Next, message recording created in chat will be described with reference to FIG.
As shown in FIG. 3, in the message record, a plurality of messages transmitted by a plurality of people are displayed in time series. In the illustrated example, messages exchanged between two people, “Takahashi” and “Inoue” are shown.
In this message recording, message numbers (reference numbers) are assigned to each message in chronological order. For example, a message number “1” is attached to one message “Thank you for your work. Here, one message is a set of text data that the user transmits at a time.

In the example shown in FIG. 3, the date and time when the message was transmitted is attached to each message. For example, the date “2016/2/4 13:15” is added to the message “Thank you for your work.
Here, it is described that the message number and date / time are displayed as a part of the image, but the server 20 only needs to acquire (store) the message number / date / time information and not display it together with the message. May be.

FIG. 4 shows an example of the topic list.
Next, a topic list created based on message recording will be described with reference to FIG. The topic list shown in FIG. 4 is created based on the message record shown in FIG.

First, the topic list indicates topics included in the message record and a period during which messages about the topic are transmitted and received. The topic list shown in FIG. 4 includes a topic name, a start message number, and an end message number.
Here, the topic name is a name given to a topic mentioned in one or a plurality of continuous messages. The start message number indicates a message number at which each topic is started. The end message number indicates the message number at which each topic ends.

  For example, in the topic about “SST test script” in FIG. 4, “start speech number” is “3” and “end speech number” is “7”. This indicates that messages from the statement number “3” to the statement number “7” in FIG. 3 are topics related to the “SST test script”.

  To explain further, from the message record of FIG. 3, “Takahashi” with message number “3”, “It is a matter of test script, but test 15 is covered with 14, so can I erase it?” The message regarding “SST test script” is from “Takahashi” of the message number “7” to “OK”.

  Note that a change in the topic list indicates that the topic has changed. In the example shown in FIG. 4, when the comment number changes from “7” to “8”, the line changes from “SST test script” to “XY Inc. case”. That is, the message number “8” indicates that the topic has changed. Here, looking at the message record of FIG. 3, it is shown that the topic has changed with the message “Is XY company in the deadline this week?” Sent by “Takahashi” with the comment number “8”. In this example, the message with the message number “8” becomes a turning point.

  In the present embodiment, the minutes are created based on the topic list as shown in FIG. For example, based on the start message number shown in FIG. 4, the message record is divided and the minutes are created. To explain further, in FIG. 4, the message of each minutes starts from the “start speech number” of “1”, “3”, “8”, “11”.

<Operation Example of Server 20>
<Topic list creation operation>
FIG. 5 is a flowchart showing an example of a topic list creation operation by the server 20.
Next, an example of a topic list creation operation by the server 20 will be described with reference to FIGS.
First, as described above, the turning point estimation operation is performed in two modes in the present embodiment. In the first aspect, a turning point is estimated from a file operation performed by a user (chat participant) executed during a period in which messages are exchanged by chat (chat execution period). In the second aspect, the turning point is estimated from the words (person names, words) included in each message.

In the following, an example of operation for creating the topic list shown in FIG. 4 based on the turning point estimated by combining the above two modes from the message record shown in FIG. 3 will be described in detail.
Here, it is assumed that the exchange of messages by chat has ended. The message record as shown in FIG. 3 is stored in the communication history storage unit 223. In the initial state, each item (topic name, etc.) in the topic list as shown in FIG. 4 is not filled, and the topic list format (storage area) is secured.

First, the topic list creation unit 203 of the server 20 puts a start message number in the youngest column in the topic list in response to an operation by the user (step 501). That is, “1” is input as the “starting speech number”. At this time, “topic name” and “end speech number” in the line where “1” is input as “start speech number” are not input (blank).
Next, the file operation estimation unit 211 of the topic conversion estimation unit 210 estimates a turning point from the file operation history (information) of the chat participant (step 502).

Next, the topic list creation unit 203 determines whether the topic list has a blank (step 503). That is, it is determined whether all the description columns of the topic list are filled.
If the topic list has a blank (YES in step 503), the related person extracting unit 217 of the person word estimating unit 213 extracts related persons (step S504). Further, the word extraction unit 215 of the person word estimation unit 213 extracts words included in the chat content (step S505).

Next, the person word estimation unit 213 estimates a turning point from the related person extracted by the related person extraction unit 217 and the word extracted by the word extraction unit 215 (step 506).
Next, the topic list creation unit 203 enters an end message number (step 507). That is, the “end speech number” is entered in the newest column in the topic list. In the example shown in FIG. 4, “12” is entered in the “end speech number” field in the “ArcSuite Steering Committee”.
On the other hand, when the topic list has no blank (NO in step 503), the process flow ends.
From the above, each item of the topic list is filled as shown in FIG.

<File operation guessing action>
FIG. 6 is a flowchart showing a topic change estimation operation based on the file operation history.
Next, with reference to FIG. 6, an operation example of topic change estimation based on the file operation history by the server 20 will be described.
Note that the topic change estimation operation example described here is an example of the operation of estimating the turning point from the file operation history by the chat participant in step 502 of FIG.

First, the file operation estimation unit 211 acquires a corresponding record list (described later) based on the file operation history stored in the file operation history storage unit 221 (step 601).
Then, the file operation estimation unit 211 determines whether there is a time when a plurality of chat participants are operating the same file in the corresponding record list (step 602).

  The topic list creation unit 203 adds (fills in) the topic list based on the information regarding the file operation that was operated when the same file was operated (step 503). That is, the topic list creation unit 203 enters “topic name”, “start speech number”, and “end speech number” in the corresponding fields (details will be described later).

<Person word guessing action>
FIG. 7 is a flowchart showing the topic change guessing operation based on the person word included in the message record.
Next, referring to FIG. 7, an example of the topic change guessing operation based on the person / word included in the message recording by the server 20 will be described.

Note that the topic change estimation operation example described here is an operation of estimating a turning point from the person extracted by the related person extraction unit 217 and the word extracted by the word extraction unit 215 in step 506 of FIG. It is an example.
Further, it is assumed that the person or word to be extracted is predetermined by a word extraction list (see FIG. 9B described later).

First, the person word estimation unit 213 identifies the related person extracted by the related person extraction unit 217 and the appearance location of the word (word) extracted by the word extraction unit 215 in the message recording (step 701).
Then, the person word guessing unit 213 determines whether the related person and the word have changed in the message recording (step 702). That is, for each occurrence of related person and word, it is determined whether the related person or word between the message including the appearance and the message before the message is different from each other.

  Then, based on the change of the related person and the word, the topic list creation unit 203 writes (adds) to the topic list (step 703). That is, the topic list creation unit 203 enters “topic name”, “start speech number”, and “end speech number” in the corresponding fields.

Next, the topic list creation unit 203 determines whether the entered column is the last topic column (step 704). That is, the topic list creation unit 203 determines whether it is the last line of the topic list.
If it is the last topic (YES in step 704), the process flow ends. On the other hand, if it is not the last topic (NO in step 704), the process proceeds to step 702.

<List of applicable records>
FIGS. 8A and 8B are diagrams for explaining the operation of acquiring the corresponding record list. Specifically, FIG. 8A shows a file operation history database, and FIG. 8B shows a corresponding record list. The file operation history database shown in FIG. 8A has a predetermined period including the time when the message included in the message record shown in FIG. 3 is transmitted and received.

Next, an operation for acquiring a corresponding record list will be described with reference to FIGS. 8 (a) and 8 (b).
First, as shown in FIG. 8A, the file operation history database is a database in which the user's file operation history is stored, and sentences executed by the clients (first client 11 to third client 13). Stores the history of operating files such as creation software, spreadsheet software, and schedule management software. In the illustrated example, the sentence creation software, spreadsheet software, and schedule management software operate with the same application as the chat.

In the illustrated example, the file operation history database includes an operation date and time, an operation file name, an operation type, and an operator (user name).
The history of operating this file is acquired by the operation information acquisition unit 201 (see FIG. 2) as described above and stored in the file operation history storage unit 221 (see FIG. 2).

  Now, the corresponding record list shown in FIG. 8B is extracted from the file operation history shown in FIG. Specifically, the file operation estimation unit 211 (see FIG. 2) extracts a list of files operated by chat participants during the chat execution period, that is, between the chat start time and end time, from the file operation history data. To do.

Here, in the example shown in FIG. 3, the chat is executed in a period from “2016/2/4 13:15” to “2016/2/4 13:38”. Then, in the file operation history database in FIG. 8A, the files operated by the chat participants executed during this period become the corresponding record list shown in FIG. 8B.
That is, the operation history (see parentheses in the figure) from “2016/2/4 13:20” to “2016/2/4 13:30” in FIG. 8A becomes the corresponding record list.

<Word extraction list>
FIGS. 9A and 9B are diagrams for explaining the operation of acquiring the word extraction list. Specifically, FIG. 9A shows a communication history database, and FIG. 9B shows a word extraction list.
Next, an operation for acquiring a word extraction list will be described with reference to FIGS. 9 (a) and 9 (b).

  First, as shown in FIG. 9A, the communication history database is a database in which the history of communication during a predetermined period (for example, from the present to half a year ago) is managed. For example, the history of communication between users, such as chat, a meeting, or an email, is included. The communication between users does not have to be via the network 40 (see FIG. 1).

  The communication history database shown in FIG. 9A is stored in the communication history storage unit 223 (see FIG. 2). The communication history database relates to communication executed before the time when the message included in the message record shown in FIG. In other words, the word extraction list extracted from the communication history database is created based on the history of other communications executed before the communication related to the message record to be divided.

  The communication history database in the illustrated example includes a communication date and time, a communication type, a communication title, and a participant. That is, the communication history database can be regarded as including keywords and file names in the past communication history.

  Now, a word extraction list shown in FIG. 9B is created from the communication history database shown in FIG. Note that the word extraction list in the illustrated example includes related persons and words related to the related persons.

First, the person word guessing unit 213 extracts chat participant information in the message recording shown in FIG. Then, in the communication history database shown in FIG. 9A, a communication history in which all chat participants are participating is extracted.
Then, in the extracted communication history, communication related persons not included in the chat participants are extracted, and a list of the related persons is created. That is, in the communication in which all chat participants have participated in the past, when there is a related person other than the chat participant, the related person is specified and included in the list.

  For example, in the example of message recording shown in FIG. 3, the chat participants are “Takahashi” and “Inoue”. Then, the communication in which two persons “Takahashi” and “Inoue” participate is specified from the communication history database shown in FIG. And the communication which has an interested party other than these 2 persons is further specified (refer the arrow in a figure). From this identified communication, the name of the person concerned and the noun part of the communication title are extracted (extracted).

  In this example, “Tanaka” and “Sato” are specified as the names of persons concerned. Further, “ArcSuite Steering Association”, “SST Test Script”, and “XY Corporation Project” are specified as noun parts (words) of the communication title. Then, as shown in FIG. 9B, a list of each identified person name and each word is listed as a word extraction list. This word extraction list can be regarded as a list in which related persons and words are linked.

In the example shown in FIG. 9B, “Tanaka” participates in the communication in which “ArcSuite Management Committee” has become a hot topic. In addition, “Sato” participates in the communication that became a hot topic of “ArcSuite Steering Committee”, “SST Test Script”, and “XY Corporation Project”.
Although details will be described later, a turning point in message recording is estimated using a person and a word included in the word extraction list shown in FIG.

  In addition, when words (keywords) appear frequently in message recording, it is relatively easy to grasp the turning point, but by using the word extraction list shown in FIG. Even if only the name comes out, it is possible to guess what topic is being discussed by the name of the person.

<Specific example of topic list creation operation>
FIGS. 10A to 10D are diagrams for explaining a specific example of the topic list creation operation by the server 20. FIG. 10A shows a corresponding record list, which has the same contents as the corresponding record list shown in FIG. FIG. 10B shows message recording, which has the same contents as the message recording shown in FIG. FIG. 10C shows word extraction, which has the same contents as the word extraction list shown in FIG. 9B. FIG. 10D shows a topic list, which has the same contents as the topic list shown in FIG.

Next, a specific example of the topic list creation operation by the server 20 will be described with reference to FIGS. 5 and 10A to 10D.
Here, it is assumed that the exchange of messages by chat has ended. In addition, it is assumed that the corresponding record list, message record, and word extraction list shown in FIGS. In addition, each item of the topic list shown in FIG. 10D is not filled, and the topic list format is secured. An operation from this state until the topic list is filled as shown in FIG. 10D, in other words, until the topic list is completed will be described below.

  First, the topic list creation unit 203 of the server 20 puts the start message number “1” (see reference numeral N1 in FIG. 10D) in the youngest column in the topic list in response to an operation by the user (see FIG. 5 above). Step 501).

  Next, as described above with reference to FIG. 6, the topic list creation unit 203 adds information on the file to the topic list when a plurality of chat participants operate the same file in the corresponding record list. To do.

  At this time, in the corresponding record list shown in FIG. 10A, the operation date / time of the file is included, but there is no message number information. Therefore, it is necessary to associate the operation date / time with the message number. On the other hand, each message shown in FIG. 10B includes information on the date and time of transmission (date and time of speech). Therefore, the utterance number of the utterance with the date and time closest to the date and time when the file was opened (operated) is the start utterance number. In the illustrated example, the message that is the closest message after the oldest time “2016/2/4 13:20” (see the symbol N2 in FIG. 10A) when the “SST test script” is opened (OPEN). The number “3” is entered as the start message number (see the symbol N3 in FIG. 10D).

  If there is already a topic list and the end message number has not been entered, the number immediately before the entered start message number is set as the end message number of the previous topic. In the illustrated example, the end message number “2” is entered (see reference numeral N4 in FIG. 10D). The topic list creation unit 203 also inputs the file name to the topic name of the line where “3” is entered as the start message number. In the illustrated example, “SST test script” is entered (see reference numeral N5 in FIG. 10B).

  Unlike the example shown in the figure, when there is a record that the file is closed (CLOSE) in the corresponding record list, the statement number of the statement with the closest date and time after the file was closed ends. It may be a speech number. Further, when there are a plurality of file operations, each record may be added. If there is no continuous file operation record corresponding to the chat message record, the message number is not filled or the message number is not continuous.

  Next, the topic list creation unit 203 determines whether the topic list has a blank (see step 503 in FIG. 5). At this time, not only the development word number and end speech number in each line of the topic list, but also whether there is a blank period (speech number omission or omission) in the speech number is checked. Then, a blank in the topic list or a portion where a speech number is missing is a target of turning point estimation based on a person / word (see step 506 in FIG. 5).

  Next, as described in step 504 in FIG. 5, in the topic list creation unit 203, the related person extraction unit 217 of the person word estimation unit 213 extracts related persons. Further, as described in step 505 of FIG. 5 above, the word extraction unit 215 of the person word estimation unit 213 extracts words included in the chat content. In the illustrated example, it is confirmed whether “related person” and “word” included in the word extraction list shown in FIG. 10C appear in the message record shown in FIG. Note that the message to be confirmed here is a message in which blanks in the topic list and missing speech numbers exist.

Next, it is confirmed from the identified related person and word whether the topic of the message has changed (changed) from the previous message.
In the example shown in the figure, when one of “related person” and “word” changes in the group of “related person” and “word” shown in the word extraction list shown in FIG. Judge that has changed.

  Specifically, when the specified word changes, it is determined that the topic has changed. If the identified person is a person different from the person related to the topic of the previous message, it is determined that the topic has changed. Also, if the identified word is the same and the person matches the person related to the topic of the previous message, it is determined that the topic has not changed.

  If a word appears next to a person, the word extraction list shown in FIG. 10C is used for comparison with reference to the word paired with that person. When a word appears next to a word, comparison is made with reference to the person paired with the word in the word extraction list shown in FIG.

In the example shown in the figure, related persons and words are identified in order in messages after the speech number “4” not included in the topic list shown in FIG.
Then, “Sato” is specified in the message record message number “5” shown in FIG. 10B (see reference numeral N6 in FIG. 10B). The “word” related to this “Sato” is confirmed in the word extraction list shown in FIG. 10C (see reference numeral N7 in FIG. 10C). Here, “Sato” of “Related Person” is described in the same row as “SST Test Script” of “Word” (see reference numeral N8 in FIG. 10C). Since this “SST test script” matches the “SST test script” (see reference numeral N5), which is the topic name immediately before in FIG. 10D, it is determined that the same topic continues. That is, the message with message number “4” in FIG. 10B is not recognized as a turning point.

Next, the same processing is performed for the related person and word specified after “Sato” in the comment number “5”. Then, in the message record message number “8” shown in FIG. 10B, “XY company” is specified (see reference numeral N9 in FIG. 10B). The “XY company” does not exactly match the “XY corporation case” (see reference numeral N10 in FIG. 10C) included in the word extraction list shown in FIG. It is recognized as being within the range of fluctuation.
The word “XY company”, that is, “XY corporation case” is a word different from “SST test script” (see reference numeral N8 in FIG. 10C). Therefore, it is determined that the topic has changed.

  Then, in the topic list shown in FIG. 10D, the speech number “8” is entered as the start speech number (see reference numeral N11 in FIG. 10D). Also, an end message number “7” is entered (see reference numeral N12 in FIG. 10D). Then, the topic list creation unit 203 inputs the identified word as the topic name in the line where “8” is entered as the start message number. In the illustrated example, “XY Co., Ltd. project” is entered (see reference numeral N13 in FIG. 10D).

  Next, “Tanaka” is specified in the message number “11” of the message record shown in FIG. 10B after the “XY Co., Ltd. case” in the message number “8” (reference number in FIG. 10B). N14). This related person is included in the word extraction list shown in FIG. 10C (see reference numeral N15 in FIG. 10C), but is “XY Co., Ltd.” (see reference numeral N10 in FIG. 10C). Not paired with. Therefore, it is determined that the topic has changed.

  Then, in the topic list shown in FIG. 10D, the speech number “11” is entered as the start speech number (see reference numeral N15 in FIG. 10D). Also, the end message number “10” is entered (see reference numeral N16 in FIG. 10D). Then, the topic list creation unit 203 inputs a word related to “Tanaka” to the topic name in the line where “11” is entered as the start message number (see reference numeral N17 in FIG. 10D). In the illustrated example, “ArcSuite Steering Committee” is entered (see reference numeral N18 in FIG. 10D).

  Next, as described in step 507 in FIG. 5, the topic list creation unit 203 enters an end message number. That is, the “end speech number” is entered in the newest column in the topic list. In the example shown in FIG. 4, “12” is entered in the “end speech number” field in the “ArcSuite Steering Committee” (see symbol N19 in FIG. 10D).

<User interface example 1>
FIG. 11 is a diagram for explaining a configuration example of a user interface for creating minutes.
Next, a configuration example of a user interface for creating minutes will be described with reference to FIG.

  As illustrated in FIG. 11, the user interface is displayed on the first client 11 to the third client 13 by the transmission unit 207. This user interface is an image displayed on the image display unit 101 (for example, a liquid crystal screen) of clients (first client 11 to third client 13) constituting the client group 10.

The user interface in this example includes a chat image 102 including a message, a text input box 103 for inputting a message, a send button 104 for transmitting a message input to the text input box 103, a voice phone, a video phone, a camera, and the like. Function buttons 105 to 107 and a minutes creation button 108.
When the chat is finished, the user selects (for example, clicks) the minutes creation button 108 to create the minutes. That is, as described with reference to FIG. 10, the message record is divided while detecting the turning point in the message record.

<User interface example 2>
FIGS. 12A and 12B are diagrams for explaining a configuration example of the user interface.
Next, a configuration example of a user interface in the minutes sending operation will be described with reference to FIG.
As described above with reference to FIG. 11, the message record is divided by the user selecting the minutes creation button 108.

  Then, as shown in FIG. 12A, an image (user interface) showing the minutes as a divided message record is displayed on the image display unit 101. In the illustrated example, a first text image 111, a second text image 113, and a third text image 115 are displayed. These first text image 111 to third text image 115 each include a display of the topic name of the topic list.

  Here, the first text image 111 is “SST test script.txt”, that is, text data with a topic name “SST test script”. This text data is data corresponding to the line of “SST test script” in FIG. 10D, and includes messages from “speech number” “3” to “speech number” “7”.

  Similarly, the second text image 113 is “XY Corporation.txt”. This text data corresponds to the line “XY Corporation” in FIG. Then, a file name of “XY Corporation.txt” is added, and a message from “8” to “10” is included.

  The third text image 115 is “ArcSuite Steering Association.txt”. This text data corresponds to the line “ArcSuite Steering Association” in FIG. 10D, and includes messages from “11” to “12”.

  In FIG. 10D, when there is no topic name in the topic list (blank), the first text image 111 or the like displays a character string indicating untitled (for example, “Untitled.txt”). The

  Here, when the title is untitled, a display may be made so that the user can recognize that the title is untitled. For example, when the first text image 111 is untitled, the display is different from the second text image 113 and the third text image 115 which are other text images, such as coloring the first text image 111 or blinking the first text image 111. The user may be able to easily grasp.

  Now, as shown in FIG. 12 (a), the mail is sent by the user to the parties involved in each message while using the image showing the divided message record (minutes). At this time, a mail is created according to the mail template information held by the transmission unit 207 (see FIG. 2).

  Here, at the time of mail transmission by the user, if the related party specified above is not included in the mail transmission destination, an assist screen for adding the related person may be displayed. For example, an image shown in FIG. 12B is displayed by pop-up or the like. In the example shown in the figure, an inquiry image 114 that displays the contents of an inquiry to the user and an instruction image 116 that receives an instruction from the user are displayed on the image display unit 101.

  In this example, it is assumed that the user did not include “Sato” as the destination of the mail to which the text data of “SST test script” is attached. On the other hand, in the word extraction list shown in FIG. 10C, “SST test script” and “Sato” are associated with each other.

  Therefore, the transmission unit 207 displays an inquiry image 114 inquiring whether or not it is necessary to send the text data of the “SST test script” to “Sato”. Then, the user selects either “YES” or “NO” in the instruction image 116 and instructs whether or not to send it to “Sato”. As a result, the omission of message recording by the user is suppressed. In addition, although displayed as “Mr. Sato” here, the display mode is not particularly limited as long as it is information related to “Sato”. For example, an e-mail address of “Sato” may be displayed.

In the present embodiment, as described above, when the contents of the chat are made into minutes, the result divided for each topic is obtained. This suppresses the occurrence of security problems by sending chat content to a person who does not need to send it. Also, by selecting a mail destination, it is possible to prevent sending unnecessary mail.
In addition, message records containing various topics are divided at turning points, so if you want to expand to several parties, or if you want to hide some topics from some parties, Forms can be used.

<Hardware configuration example of server 20>
FIG. 13 is a diagram illustrating a hardware configuration example of the server 20.
The server 20 is realized by a computer device as described above.
As illustrated in FIG. 13, the server 20 includes a CPU (Central Processing Unit) 231 that is a calculation unit, a main memory 233 and a HDD (Hard Disk Drive) 235 that are storage units. Here, the CPU 231 executes various programs such as an OS (Operating System) and application software. The main memory 233 is a storage area for storing various programs and data used for execution thereof. The HDD 235 is a storage area for storing input data for various programs, output data from the various programs, and the like.

  The server 20 also includes a communication interface (communication I / F) 237 for performing communication with the outside. Note that the program executed by the CPU 231 is stored in a storage medium such as a CD-ROM, for example, in addition to being stored in the main memory 233 in advance, or provided to the CPU 231 or the network 40 (see FIG. 1). It is also possible to provide it to the CPU 231.

<Others>
In the above description, it has been described that a turning point is detected by detecting related persons and words included in the word extraction list as shown in FIG. However, the method for detecting the turning point is not particularly limited. For example, a phrase such as a conjunction used when changing a topic such as “by the way” or “speaking” may be detected, and the turning point may be detected based on the phrase.

  In the description of FIG. 8B described above, when a plurality of chat participants are operating the same file, the information of the file is added to the topic list. However, the present invention is not limited to this. . For example, even if the materials to be operated are not the same file, it may be detected when a user operates a file held by each user with the same name, or when the same software is used.

  Moreover, the case where some chat participants are using the same file among several chat participants may be sufficient. In other words, it is only necessary to execute an operation (common operation) common to two or more chat participants, even if not all chat participants.

Further, in the description of FIG. 11 described above, it has been described that the minutes are created at the end of the chat, but the present invention is not limited to this. After the chat is finished, the minutes may be created after being stored in the storage means.
In the above description, the turning point is estimated based on the person / word after the turning point is estimated based on the file operation. However, the present invention is not limited to this. After guessing the turning point based on the person / word, the turning point may be guessed based on the file operation. Further, in a message included between a plurality of turning points based on file operations, a turning point may be estimated based on a person / word.

  In the above description, the divided message record is transmitted by mail. However, the present invention is not limited to this. For example, it suffices if a message record divided into related parties such as FAX and paper media can be transmitted to participants and related parties. Note that it is not essential to transmit the message record, and it is only necessary to grasp the turning point.

  In the above description, the chatting, that is, the detection of the turning point when exchanging text messages has been described. However, the present invention is not limited to this. Any text data indicating communication results (information exchange history) between a plurality of persons may be used. For example, the present invention is applicable to a case where a plurality of e-mails are used or a case where a conversation between users is converted into a text. is there.

Although various embodiments and modifications have been described above, the embodiments and modifications may be combined with each other.
Further, the present disclosure is not limited to the above-described embodiment, and can be implemented in various forms without departing from the gist of the present disclosure.

DESCRIPTION OF SYMBOLS 1 ... Recording division | segmentation system, 10 ... Client group, 20 ... Server, 203 ... Topic list creation part, 210 ... Topic change guess part, 211 ... File operation guess part, 213 ... Person word guess part, 205 ... Minutes creation part

Claims (8)

An acquisition unit for acquiring a message record in which a plurality of messages transmitted and received via a telecommunication line are arranged in time series among a plurality of users;
A storage unit that stores an operation history that is a history of a user operating a device connected to the telecommunication line;
The message recording based on a common operation which is an operation recorded in the operation history and is performed when a plurality of messages in the message record are transmitted and received and which is common to at least two users. A recording / dividing device. Having another storage unit that stores an information exchange history that is a history of information exchange among a plurality of parties including the plurality of users;
The recording division apparatus according to claim 1, wherein the division unit divides the message record based on the common operation and the information exchange history. Another storage unit that stores an information exchange history that is a history of information exchange among a plurality of parties including the plurality of users;
A specific unit that identifies the parties recorded in the information exchange history and words related to each information exchange;
The recording according to claim 1, wherein the dividing unit divides the message record based on at least one of the related parties and the related words included in the message record and the information exchange history. Splitting device. A set specifying unit for specifying a plurality of sets of the related person and the related word in the information exchange history;
The division unit specifies at least one of the related person and the related word included in the message record, and the set including the specified one is determined by the specified related person or the related word. 4. The recording / splitting apparatus according to claim 3, wherein, when the set is different from the group in the message before the included message, the message record is split between the previous message and the message including the one. A display unit that displays information of a person who is at least one of the parties concerned and is related to the content of the divided message record, as a transmission destination candidate for transmitting the divided message record;
5. The recording and dividing apparatus according to claim 2, further comprising a receiving unit that receives an instruction to transmit the divided message record to the related person displayed on the display unit. 6. . 6. The recording according to claim 1, wherein the dividing unit divides the message recording based on an operation of the same file by the at least two users as the common operation. Splitting device. An acquisition unit for acquiring a message record in which a plurality of messages transmitted and received via a telecommunication line are arranged in time series among a plurality of users;
In the message recording, an extraction unit that extracts a turning point at which a topic referred to in the plurality of messages has changed,
A recording / dividing apparatus comprising: a dividing unit that divides the message record based on the turning point extracted by the extracting unit. A providing unit for providing a service for transmitting and receiving a plurality of messages via a telecommunication line between a plurality of users;
An acquisition unit for acquiring a message record in which the plurality of messages are arranged in time series;
A storage unit that stores an operation history that is a history of a user operating a device connected to the telecommunication line;
The message recording based on a common operation which is an operation recorded in the operation history and is performed when a plurality of messages in the message record are transmitted and received and which is common to at least two users. And a dividing unit for dividing the recording medium.

Images