JP2005025642A - Message processing device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To process sensitive information while shielding from electronically transmitted messages such as inquiries and complaints. <P>SOLUTION: A customer transmits a message to a consultation center clerk (X1). The message is relayed by a message processor 10, and each process of sensitivity expression extraction (X2), unique expression extraction (X3), ranking (X4), destination candidate discrimination (X5), and markup process (X6) is executed thereto. The message is transferred to a person in charge of development through a consultation center clerk. At this time, the message processor 10 performs shielding process of a unique expression (X9). An answer message is relayed by the message processor 10 and transmitted to the consultation center clerk. At this time, the shielded part in the message is restored to the unique expression (X12 and X13). The consultation center clerk prepares a message for the customer and transmits it to the customer (X14). <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention is a message processing technique for analyzing and processing messages (text) such as inquiries and complaints sent electronically, such as mail transmission, HTTP transmission, FTP transmission, etc. based on the presence or absence of expressions that touch sensibility. In particular, it is intended to shield sensitive information.
[0002]
[Prior art]
How to extract important information included in inquiries and complaints from customers is a matter of life and death for companies. In recent years, such inquiries have been sent by electronic methods such as e-mails. Depending on whether the inquiries sent by such electronic methods include dissatisfaction or anger. In addition, it is necessary to distribute it to a desired destination such as a product developer (inside the same company or in the case of an OEM product, an OEM manufacturer). However, such inquiries and other messages often contain sensitive information (the contents of the inquiries are sensitive because the inquiries often expect the inquires to be kept confidential) In many cases, it is included as is), and if sent directly to the distribution destination, a problem occurs.
[0003]
In addition, as patent documents related to the present invention, there is a method in which a questionnaire response is received by e-mail, a response including a complaint sentence is detected using a complaint dictionary, and statistical processing is performed (for example, Patent Document 1). .
[Patent Document 1]
JP2000-181926
[0004]
[Problems to be solved by the invention]
The present invention has been made in view of the above circumstances, and an object thereof is to provide a message processing technique that shields and processes sensitive information from electronically transmitted messages such as inquiries and complaints. It is said.
[0005]
[Means for Solving the Problems]
According to this invention, in order to achieve the above-mentioned object, the configuration as described in the claims is adopted. Here, prior to describing the invention in detail, supplementary explanations of the claims will be given.
[0006]
That is, according to one aspect of the present invention, in order to achieve the above-described object, the message processing apparatus includes: a message receiving unit that receives a message; and a specific expression extracting unit that extracts a specific expression included in the received message. A replacement means for replacing a predetermined specific expression in the received message with another notation in which the content of the specific expression is unknown; a message for transferring the message in which the specific expression is replaced with the other expression to a predetermined destination Transfer means; and restoration means for restoring the other notations included in the message transferred from the destination into corresponding specific expressions are provided.
[0007]
The specific expression is an expression (also referred to as a specific name) that represents a specific thing such as a person name, a place name, an organization name, a date, and a price.
[0008]
In this configuration, since the predetermined specific expression is replaced with another notation by the replacing means and transferred to the predetermined destination, no problem occurs in the destination even if the specific expression is sensitive information itself. The predetermined specific expression may be, for example, all the extracted specific expressions, or may be selected according to a predetermined standard from the extracted specific expressions. As other notations in which the content of the unique expression is unknown, various things such as a serial number, a list of meaningless symbols, and combinations thereof can be adopted. When a message is expressed using XML or HTML, an identification number for specifying the specific expression may be embedded by a tag, and a meaningless symbol such as a black circle or a white circle may be expressed.
[0009]
The message returned from the destination includes the entire text of the original message or a part thereof, and the other notation in the returned message is restored to the original unique expression by the restoring means. The returned message may be directly transferred from the destination or may be transferred via any relay point.
[0010]
The message is basically a text message such as an inquiry or complaint, but may be voice data. Voice data is converted to text. The message is transferred by, for example, a mail transfer method, an HTTP transfer method, an FTP transfer method, or the like, but is not limited thereto.
[0011]
According to another aspect of the present invention, the message processing device includes: a message receiving unit that receives a message; a sensitivity expression extracting unit that extracts a feeling touched in the received message; and the received message Specific expression extraction means for extracting the specific expression included in the message; importance level determination means for determining the importance level based on the extracted expression touching the sensitivity; ranking the messages based on the determined importance level And means for highlighting and presenting the message touching the sensibility and the specific expression; and replacing means for replacing the predetermined specific expression in the received message with another expression whose content of the specific expression is unknown. And message transfer means for transferring the message obtained by replacing the unique expression with the other notation to a predetermined destination The are to be provided.
[0012]
In this configuration, the importance level determination means is used to determine the importance level of the message based on the expression touching the sensibility included in the message, the ranking is performed, and the expression that touches the sensitivity and the specific expression are highlighted. Processing according to the importance of the message can be reliably performed.
[0013]
Note that highlighting broadly means that the part is displayed in a mode different from other parts, and can be performed in various modes such as highlighting and underlining.
[0014]
The present invention can be realized not only as an apparatus or a system but also as a method. Of course, a part of the invention can be configured as software. Of course, software products used to cause a computer to execute such software are also included in the technical scope of the present invention.
[0015]
These and other aspects of the invention are set forth in the appended claims and will be described in detail below with reference to examples.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
Examples of the present invention will be described below.
[0017]
FIG. 1 shows the basic configuration of a message processing apparatus according to an embodiment of the present invention. In this figure, the message processing apparatus 10 uses a transmission / reception unit 101 for a customer host device 20, an administrator host device 21, The respondent host device 22 and the like send and receive messages. In this embodiment, the customer sends a message such as a complaint or inquiry to the customer consultation center staff using the customer host device 20, and the customer consultation center staff sends this message to the manager host device 21, for example. It is received via the processing device 10. The Customer Consultation Center representative will request that the in-house development staff forward the message to the staff of the external partner company as appropriate. At this time, sensitive information is transferred while being concealed (hereinafter also referred to as a mask). An in-house development staff member in charge of an external partner company returns a reply message to a customer consultation center staff member via the message processing device 10 using, for example, the respondent host device 22. At this time, the hidden information is restored. The person in charge of the customer consultation center refers to the reply message and sends a predetermined message to the customer.
[0018]
The message processing apparatus 10 includes a transmission / reception unit 101, a sentiment expression extraction unit 102, a specific expression extraction unit 103, a message evaluation unit 104, a concealment unit 105, a restoration data storage unit 106, a restoration processing unit 107, and the like. Yes. The message storage unit 108 stores and manages messages exchanged between the units.
[0019]
The sensibility expression extraction unit 102 detects an expression that touches sensibility (hereinafter also referred to as sensibility expression). Specific examples thereof will be described later in detail with reference to FIGS.
[0020]
The specific expression extraction unit 103 detects a specific expression, and a specific example thereof will be described in detail later with reference to FIG.
[0021]
The message evaluation unit 104 evaluates a message based on the emotional expression and the specific expression detected by the emotional expression extraction unit 102 and the specific expression extraction unit 103 and performs a desired process. For example, as shown in FIG. In addition, it includes a ranking unit 110, a destination candidate determination unit 111, a markup unit 112, and the like.
[0022]
The ranking unit 110 refers to, for example, the ranking table 113 (see, for example, FIG. 3) to obtain the cumulative value or the highest point value of the emotional expression included in the message, and based on this, rank of each message To do. The approximate importance and urgency of the message can be determined by ranking. It is possible to control to process a message of higher rank first, change the color notation, and add a warning display.
[0023]
The destination candidate determination unit 111 determines destination candidates with reference to the destination table 114 (see, for example, FIG. 4) and presents them to the customer consultation center staff. A single destination may be presented, or a plurality of options may be presented. It is also preferable that the user (person in charge of the customer consultation center) can manually input a destination other than the candidate.
[0024]
The markup unit 112 refers to a markup table 115 (for example, see FIG. 5) to change the color attribute of the character string of the emotional expression or the unique expression. When a message is written in HTML, the font color and size are selected by a tag. Here, the corresponding portion is highlighted by changing the color attribute, but the corresponding portion can be displayed differently from the other portions in various other forms (highlighted display). For example, an underline, a bold character, an italic character, or the like may be used.
[0025]
The concealment unit 105 masks the unique expression in the message when sending a message from the administrator host device 21 to the respondent host device 22 (see FIG. 10). In this example, an identification number (three digits) is assigned to each message in the order of appearance of the unique expression, and black square marks are attached before and after that to replace the unique expression. The replacement result is stored in the restoration table of the restoration data storage unit 106 as shown in FIG. The secondary data is not limited to this.
[0026]
The restoration processing unit 107 restores the concealed portion in the message to the original unique expression when the message is forwarded from the respondent host device 22 to the administrator host device 21. The restoration is performed using the restoration table (FIG. 6) of the restoration data storage unit 106.
[0027]
FIG. 7 shows the operation of the embodiment. In this figure, the customer transmits a message from the customer host device 20 to the customer consultation center staff (X1). This message contains a complaint as shown in FIG. 8, for example. This message is relayed by the message processing device 10, and each process of sensitivity expression extraction (X2), specific expression extraction (X3), ranking (X4), destination candidate determination (X5), and markup processing (X6) is executed. Thereafter, it is received by the person in charge of the customer consultation center using, for example, the administrator host device 21 (X7). The marked up state is as shown in FIG. 9, for example. The person in charge of the customer consultation center transfers the message to the person in charge of development using, for example, the respondent host device 22 according to the destination information according to the destination candidates or manually input by himself / herself. At the time of this transfer, the message processing device 10 relays the message (X8), performs the concealment process of the specific expression (X9), and transfers it to the developer (X10). The message after concealment is, for example, as shown in FIG. The person in charge of development browses the message (FIG. 10) in which the unique expression is concealed, for example, on the respondent host device 22, and creates a reply message (FIG. 11). At this time, no problem arises because the sensitive specific expressions are hidden. The reply message is relayed through the message processing apparatus 10 and transmitted to the customer consultation center staff (X11). The hidden part in this message is restored to a specific expression (X12, X13). The message after restoration is as shown in FIG. Based on this message, the customer consultation center staff prepares a message for the customer (FIG. 13) and transmits it to the customer (X14).
[0028]
Note that when the hidden expression is masked uniformly or in accordance with a predetermined rule by the concealment unit 105, it is possible that what is not originally masked is masked, and conversely, what is desired to be masked is not masked. . In order to deal with such a case, as shown in FIG. 14, for example, a control instruction is given so that the user (person in charge of the customer consultation center) can cancel the concealment or select a new concealment target. May be input from the administrator host device 21 to the hidden shielding unit 105 via the hidden shielding control input unit 116. For example, when masking is automatically performed as shown in FIG. 15, the purpose may not be achieved if the product name is unknown, and it is not desirable to transmit an expression such as “Let's go wrong!” And In that case, a control signal of “003 concealment cancellation” is sent from the administrator host device 21 to the concealment unit 105 via the concealment control input unit 107. Also, specify the area of “Let's make a difference!” And give “Hidden Intercept”. As a result of such an instruction, as shown in FIG. 16, the product name “Sashisuseso” is written, and “Need to change!” Is registered as a mask object in the restoration data storage unit 106 as “E001”, Masked up.
[0029]
FIG. 17 shows a configuration example in which the above-described embodiment is realized using a mail server. In FIG. 17, portions corresponding to those in FIG.
[0030]
In FIG. 17, a POP server 30, an SMTP server 31, an administrator host device 21, a router 32, and the like are connected to the LAN 33. A message processing unit 301 is provided as a front-end module of the POP server 30. This message processing unit 301 corresponds to the message processing device 10 in FIG. 1, and is based on the rule specified by the source / destination pair of the input / output message of the POP server 30 in the message processing device 10 in FIG. Perform the corresponding process. For example, for mail (message) sent from the customer host device 20 (another host device such as a respondent host device) to the administrator host device 21, sensitivity expression extraction, unique expression extraction, ranking, destination Candidate selection, markup processing, etc. Concealment processing is performed on mail sent from the administrator host device 21 to the respondent host device 22. The mail sent from the respondent host device 22 to the administrator host device 21 is restored.
[0031]
In this example, the customer host device 20 and the respondent host device 22 are connected to the message processing system (mail system) using the Internet 33. Of course, communication resources other than the Internet 33 can be used. Yes, the administrator host device 21 may be connected to the Internet 33.
[0032]
FIG. 18 shows another configuration example in which the above-described embodiment is realized mainly using a web server. In FIG. 18, portions corresponding to FIG. 1 or FIG.
[0033]
In FIG. 18, a web server 34 and an application server 35 are used. The user uses the web server 34 to access the system by specifying a URL, for example. Users can send and receive messages using web pages. The message processing application 351 of the application server 35 performs the same processing as that of the message processing device 10 of FIG. The web generation application 352 dynamically generates a necessary web page. Of course, a static web page may be prepared and used in advance. Further, in order to notify the user of the URL for access, a mail in which the URL is embedded may be generated by the application server 35 and sent to a predetermined destination by the SMTP server 31. Further, the CGI program of the web server 34 may be used instead of the application server 35.
[0034]
Next, a specific example of the sensibility expression extraction unit 102 in FIG. 1 will be described.
[0035]
First, the grammatical terms used below are organized.
(1) Sentences: Independent words only, or start with independent words and end with attached words.
(2) Autonomous words: nouns, pronouns, verbs, adjectives, adjective verbs, conjunctions, impression verbs, conjunctions, adverbs.
(3) Adjunct: particle, auxiliary verb.
(4) Precautions: Verbs, adjectives, adjective verbs.
(5) Body language: Refers to nouns and pronouns.
(6) Dependency (modification / substitution): When clauses are connected in meaning, it is said that the previous clause relates to the subsequent clause, and the subsequent clause receives the previous clause. The person who receives it is a modification, and the person who receives it is a modification.
[0036]
FIG. 19 shows an entire Kansei expression extracting apparatus according to an embodiment of the present invention. In this figure, the Kansei expression extracting apparatus includes a text input unit 200, an important phrase specifying unit 201, an important continuous clause specifying unit 202, The sentence type determination unit 203, the sensitivity expression determination unit 204, the extraction result storage unit 205, the user interface unit 206, the sentence type determination information storage unit 207, and the sensitivity expression determination information storage unit 208 are configured.
[0037]
The text input unit 200 inputs text one sentence at a time from data files, character recognition results, speech recognition results, and the like.
[0038]
The important phrase specifying unit 201 specifies a phrase that seems to be the most important in the input sentence. A method of selecting a phrase at the end of a sentence by performing morphological analysis and dependency analysis can be adopted. Furthermore, a method of extracting a plurality of prescriptive phrases can also be adopted. Furthermore, a method of assigning importance ranking between them can also be adopted.
[0039]
The important phrase specifying unit 201 includes, for example, as shown in FIG. 20, a morphological analysis unit 2010, a syntax analysis unit 2011, a dependency analysis unit 2012, a main phrase selection unit 2013, and the like. The morpheme analysis unit 2010 performs morpheme analysis as shown in FIG. 21, for example. The syntax analysis unit 2011 outputs a syntax analysis tree as shown in FIG. 22, for example. The main phrase selector 2013 selects, for example, a phrase at the end of the sentence as a main phrase.
[0040]
Returning to FIG. The important continuous clause specifying unit 202 specifies the continuous clause evaluated as the most important among the continuous clauses having a dependency relationship with the clause specified by the important phrase specifying unit 201 as the important consecutive clause. For example, in the dependency analysis, a continuous clause that is close to the central phrase is selected. Alternatively, importance is assigned in the order of “ha”, “ga”, “wo”, and “ni”, and selection can be made based on the importance. You may give importance based on proximity and the kind of case.
[0041]
The sentence type determining unit 203 determines the type of the sentence by referring to the sentence type determining information storage unit 207 for the sentence including the important phrase and the important continuous phrase. The sentence type determination information in the sentence type determination information storage unit 207 is, for example, information of a prescription classification table as shown in FIG. In the example shown in FIG. 23, a type and a sensitivity value are held for each combination of a precaution and a body. For example, in the case of a sentence “can be used”, the top entry in the example of FIG. 23 corresponds to the type of “impression”. It can also be seen that the sensitivity value is “+2”. In the example of FIG. 23, a type is assigned to the broader concept word. That is, a high-level concept word is assigned to each prescription, and the type of sentence including the prescription is determined based on this. Of course, a type may be assigned to each predicate without interposing a broader concept word. Types include, for example, opinions, impressions, backgrounds, questions, requests, situation explanations, complaints, and the like. Of course, the type may be selected depending on what purpose is used.
[0042]
The emotional expression determination unit 204 determines the emotional expression using the emotional expression determination information stored in the emotional expression determination information storage unit 208. For example, a sentence including a predetermined type of predicate is determined as an emotional expression. Then, for example, the sensitivity values of the prescription classification table shown in FIG. 23 can be output together. A flag or the like indicating whether the expression is a Kansei expression can be assigned independently from the sentence type, and it can be determined whether the expression is a Kansei expression. If there is a sensitivity value, it may be determined as a sensitivity expression. The sensitivity value represents, for example, a positive / negative degree. In the example of FIG. 23, a five-level evaluation (-2, -1, 0, 1, 2) is adopted.
[0043]
The determination result of the sentence type determination unit 203 and the determination result of the sentiment expression determination unit 204 are held in the extraction result storage unit 205. It is possible to acquire various information based on the type and the sensitivity value of the sensitivity expression and acquire or quantify important information included in the original text.
[0044]
The user interface unit 206 receives various inputs from the user, and specifically includes a display, a keyboard, a mouse device, and the like. In this example, input is received from a remote host device.
[0045]
Next, the operation of the configuration example of the sensitivity expression extraction unit 102 will be described.
[0046]
FIG. 24 shows the operation of the configuration example. In this figure, the following processing is performed.
[0047]
[Step S10]: Text sentences are input one by one.
[Step S11]: Important phrases in the sentence are specified.
[Step S12]: An important continuous clause corresponding to the important clause is identified.
[Step S13]: A prescriptive classification table is drawn.
[Step S14]: If there is an entry for the corresponding word, the process proceeds to step S16. If there is no corresponding prescriptive entry, the process proceeds to step S15.
[Step S15]: A classification name (type) is input using a predetermined user interface, and a sensitivity value is input. A broad concept word may be input. The input result is registered as a new entry in the prescription classification table (FIG. 22). Thereafter, the process proceeds to step S16.
[Step S16]: The sentence type is determined according to the contents of the prescription classification table (or the contents input by the user).
[Step S17]: It is determined whether or not the emotional expression is in accordance with the sentence type or the like. When the sensitivity value is acquired, the sensitivity value is output.
[0048]
Next, a specific example of the specific expression extraction unit 103 in FIG. 1 will be described.
[0049]
FIG. 25 shows a configuration example of the specific expression extraction unit 103 as a whole. In FIG. 25, the specific expression extraction unit 103 includes a text input unit 400, a morpheme analysis unit 401, a morpheme analysis dictionary storage unit 402, and specific expression information. It includes an analysis unit 403, a specific expression component candidate dictionary storage unit 404, a specific expression specifying unit 405, a specific expression extraction rule storage unit 406, an output unit 407, and the like.
[0050]
FIG. 26 illustrates processing (steps S20 to S24) performed by the specific expression extraction unit 103 in FIG.
[0051]
25 and 26, the text input unit 400 inputs Japanese text (S20). The morpheme analysis unit 401 refers to the morpheme analysis dictionary storage unit 402 and performs morpheme analysis on the text (S21). The morphological analysis result is, for example, as shown in FIG. In this example, morphological analysis is performed on "Orange group of California, USA ...". The specific expression information analysis unit 403 refers to the specific expression component candidate dictionary storage unit 404 and assigns a feature (an attribute related to the specific expression component candidate) to the morphological analysis result (S22). The specific expression component candidate dictionary storage unit 404 stores, for example, a specific expression component candidate dictionary including information associating morphemes with positions in the specific expressions formed by the morphemes as illustrated in FIG. If the proper expression is “Japan”, “Japan” is the leftmost morpheme and “country” is the rightmost morpheme. In the case of the “Constitution of Japan”, “Japan” is the leftmost morpheme, “Country” is the intermediate morpheme, and “Constitution” is the rightmost morpheme. In this case, “country” is both a leftmost morpheme and an intermediate morpheme.
[0052]
The features given to the morpheme by the specific expression information analysis unit 403 are as shown in FIG. 29, for example. In this example, the feature is determined from the attribute of the specific expression itself and the attribute related to the position of the morpheme constituting the specific expression.
[0053]
If the morpheme of the morpheme analysis result is the “word” itself, “leftmost” morpheme itself, “intermediate” morpheme itself, or “rightmost” morpheme itself as indicated by arrow A in FIG. Give each element a unique feature. For example, if it is the “right end” morpheme of “place” (for example, “country” of “Japan”), “prb” is assigned.
[0054]
In addition, when a part of the morpheme of the morpheme analysis result matches the “word”, “left end” morpheme, “intermediate” morpheme, and “right end” morpheme of the specific expression, the feature as shown by an arrow B in FIG. Give. For example, if a morphological analysis of “... Korean soldiers ...” was performed to obtain a morpheme sequence of “Korea” and “military”, the “army” in the morpheme “military” was the “right end” of “organization”. Because it can be a morpheme, left The position information of “orbl” is assigned to “military”. In the case of adopting a description method in which a feature is individually assigned to a part of a morpheme such as “military” in “military”, a feature related to a part of the morpheme may be assigned.
[0055]
In this way, the morpheme of the morpheme analysis result or a part thereof is analyzed with reference to the specific expression component candidate dictionary, and the feature is assigned to the morpheme.
[0056]
FIG. 30 shows an example in which features are assigned to the morphological analysis results of FIG. In this example, the underlined part is newly assigned as a feature.
[0057]
The specific expression specifying unit 405 refers to the specific expression specifying rule storage unit 406 and specifies a morpheme string that matches the rule as a specific expression (S23).
[0058]
The specific identification rules (also referred to as chunking rules) in the specific expression specific rule storage unit 406 are as shown in FIG. Is done. Then, as shown in FIG. 32, a morpheme string having a specific expression attribute is extracted as a specific expression. In the example of FIG. 32, the item having the attribute “location” is extracted as “location ()”. In FIG. 31, “*” indicates that the morpheme immediately before is repeated 0 or more times, and “+” indicates that the morpheme immediately before is repeated 0 or more times. “?” Represents an arbitrary morpheme.
[0059]
FIG. 33 shows an example in which a specific expression and its attributes are specified by applying a specific expression specifying rule to the morphological analysis result of “Orange County in California, USA ...”. FIG. The example which extracted expression is shown. In this example, the morpheme of the morpheme analysis result is the specific expression component candidate (left end, middle, right end, word) itself.
[0060]
FIG. 35 shows an example in which a part of the morpheme obtained as a result of the morpheme analysis constitutes a unique eaves component candidate. In this example, as described above, the left part of “military” in the morpheme sequence of “Korea” and “military” included in the morphological analysis result of “... Korean soldier ...” can be the rightmost morpheme. So “military” “organization” “ left "Orbl" which is a feature including the right end is given.
[0061]
The extracted specific expression is output from the output unit 407 after being emphasized or concealed in the text. For output, various forms such as display, printing, mail transmission, and voice output can be used. The text is output as text or the like with specific expression information to various processing devices in the subsequent stage.
[0062]
As another configuration example of the specific expression extraction unit 103, the one proposed by the applicant in Japanese Patent Application No. 2002-266227 may be used.
[0063]
The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the invention. For example, in the above-described embodiment, the specific expression is mainly hidden, but the present invention is not limited to this and can be applied to the hidden / restoration of various expressions.
[0064]
【The invention's effect】
As described above, according to the present invention, sensitive information can be shielded and processed from electronically transmitted messages such as inquiries and complaints.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating the basic configuration of an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration example of a message evaluation unit 104 in FIG.
3 is a diagram illustrating a configuration example of a ranking unit 110 in FIG.
4 is a diagram for explaining a configuration example of a destination candidate determination unit 111 in FIG. 2; FIG.
5 is a diagram illustrating a configuration example of a markup unit 112 in FIG.
6 is a diagram illustrating processing of the restoration processing unit 107 in FIG. 1. FIG.
FIG. 7 is a diagram for explaining the operation of the embodiment of FIG. 1;
FIG. 8 is a diagram for explaining an example of a mode of a message (message sent from a customer) in the embodiment of FIG. 1;
9 is a diagram for explaining an example of a mode of a message (message being processed by a customer consultation center staff) in the embodiment of FIG. 1; FIG.
10 is a diagram for explaining an example of a mode of a message (message sent to a developer in charge) in the embodiment of FIG. 1; FIG.
11 is a diagram for explaining an example of a mode of a message (message sent from a developer in charge) in the embodiment of FIG. 1; FIG.
12 is a diagram for explaining an example of a mode of a message (a state in which a customer consultation center person is browsing a message sent from a development person in charge) in the embodiment of FIG. 1;
13 is a diagram for explaining an example of a mode of a message (a message sent from a customer consultation center person in charge to a customer) in the embodiment of FIG. 1;
FIG. 14 is a block diagram illustrating a modified example in which concealment and shielding can be controlled.
FIG. 15 is a diagram for explaining an example of a state before the concealment is controlled.
FIG. 16 is a diagram for explaining an example of a state after the concealment is controlled.
FIG. 17 is a diagram showing a configuration example in which the embodiment of FIG. 1 is mainly constructed by a mail system.
18 is a diagram showing a configuration example in which the embodiment of FIG. 1 is mainly constructed by a web system.
FIG. 19 is a block diagram illustrating a configuration example of a sensitivity expression extraction unit 102 according to the embodiment of FIG. 1;
FIG. 20 is a block diagram illustrating a configuration example of an important phrase specifying unit 201 in the above configuration example.
FIG. 21 is a diagram describing an example of morphological analysis of the above configuration example.
FIG. 22 is a diagram illustrating an example of syntax analysis of the configuration example described above.
FIG. 23 is a diagram illustrating a prescription classification table that can be used for determination of sentence type and determination of emotional expression in the above configuration example.
FIG. 24 is a flowchart illustrating the operation of the above configuration example.
FIG. 25 is a block diagram illustrating a configuration example of a specific expression extraction unit 103 in the embodiment of FIG. 1;
26 is a flowchart for explaining the operation of the configuration example of FIG. 25;
FIG. 27 is a diagram for explaining a specific expression component candidate dictionary in the configuration example of FIG. 25;
FIG. 28 is a diagram for explaining the morphological analysis result of the configuration example of FIG.
29 is a diagram illustrating features used in the configuration example in FIG. 25. FIG.
30 is a diagram for explaining a result of reflecting a feature in a morphological analysis result in the configuration example of FIG. 25;
FIG. 31 is a diagram illustrating a chunking rule in the specific expression specifying unit of the configuration example in FIG. 25;
32 is a diagram for explaining extraction rules in a specific expression specifying unit in the configuration example of FIG. 25;
33 is a diagram illustrating an example of an analysis result after application of the chunking rule in the configuration example of FIG. 25;
34 is a diagram illustrating an example of an extraction result obtained by applying the extraction rule of FIG. 32 to the analysis result of FIG.
35 is a diagram showing another example of the analysis result after application of the chunking rule in the configuration example of FIG. 25. FIG.
[Explanation of symbols]
10 Message processing device
20 Customer host device
21 Administrator host device
22 respondent host device
30 POP server
31 SMTP server
32 routers
33 Internet
34 Web server
35 Application server
101 transceiver
102 Kansei expression extraction unit
103 Named entity extraction unit
104 Message Evaluation Department
105 Concealment block
106 Data storage unit for restoration
107 Concealment control input part
107 Restoration processing unit
108 Message storage
110 Ranking section
111 Destination candidate determination unit
112 Markup Club
113 ranking table
114 Destination table
115 Markup table
116 Concealment control input part
200 Text input part
201 Important phrase identification part
202 Important consecutive clause identification part
203 sentence type determination part
204 Kansei expression determination unit
205 Extraction result storage
206 User interface section
207 Information storage unit for sentence type determination
208 Information storage unit for sensitivity expression determination
301 Message processor
351 Message Processing Application
352 Web generation application
400 Text input part
401 Morphological analyzer
402 Morphological analysis dictionary storage unit
403 Specific expression information analysis unit
404 Specific expression component candidate dictionary storage unit
405 Specific expression specifying part
406 Specific expression extraction rule storage unit
406 Specific expression specifying rule storage unit
407 Output section
2010 Morphological Analysis Department
2011 Parsing section
2012 Dependency Analysis Department
2013 Main phrase selector

Claims (9)

Message receiving means for receiving a message;
A named entity extracting means for extracting a named entity included in the received message;
Replacement means for replacing a predetermined specific expression in the received message with another notation that does not know the content of the specific expression;
Message transfer means for transferring the message obtained by replacing the specific expression with the other notation to a predetermined destination;
A message processing apparatus comprising: a restoring unit that restores the other notation included in the message transferred from the destination to a corresponding specific expression. The message processing apparatus further comprising means for discriminating the predetermined destination based on the specific expression extracted from the message. Message display means for highlighting a portion of the specific expression of the message based on the specific expression extracted from the message;
Destination candidate display means for displaying the predetermined destination candidates;
The message processing apparatus according to claim 1, further comprising selection input receiving means for receiving a selection input for selecting one destination from the displayed destination candidates. Message receiving means for receiving a message;
Replacement means for replacing the predetermined expression in the received message with another notation that does not know the content of the expression;
Association storage means for associating and storing the predetermined expression and the other expression;
Message transfer means for transferring the message obtained by replacing the predetermined expression with the other notation to a predetermined destination;
A message processing apparatus comprising: a restoring unit that restores the other notation included in the message transferred from the destination into a corresponding expression with reference to the association storage unit. Message receiving means for receiving a message;
Kansei expression extracting means for extracting an expression touching the sensitivity included in the received message,
A named entity extracting means for extracting a named entity included in the received message;
Importance determining means for determining the importance based on the extracted expression touching the sensitivity,
A presenting means that ranks the messages based on the determined importance and highlights and presents the expression touching the sensibility and the specific expression;
Replacement means for replacing a predetermined specific expression in the received message with another notation that does not know the content of the specific expression;
And a message transfer unit configured to transfer the message obtained by replacing the specific expression with the other expression to a predetermined destination. The message processing apparatus according to claim 5, wherein the importance is determined by referring to the extracted specific expression in addition to the extracted expression touching the sensitivity. The message processing apparatus according to claim 6, further comprising a restoration unit that restores the other notation included in the message transferred from the destination to a corresponding specific expression. Receiving a message;
Extracting a specific expression included in the received message;
Replacing the predetermined specific expression in the received message with another notation that does not know the content of the specific expression;
Forwarding the message with the unique representation replaced with the other notation to a predetermined destination;
Restoring the other notation included in the message transferred from the destination to a corresponding specific expression. Receiving a message;
Extracting a specific expression included in the received message;
Replacing the predetermined specific expression in the received message with another notation that does not know the content of the specific expression;
Forwarding the message with the unique representation replaced with the other notation to a predetermined destination;
A computer program for message processing, which is used for causing a computer to execute the step of restoring the other notation included in a message transferred from the destination to a corresponding specific expression.

Images