JP2000331018A - Electronic information reception equipment and electronic information processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To inform a user of only data which seems to be a topic by automatically analyzing contents of stored mail list data. SOLUTION: This electronic information reception equipment is provided with a central processing unit 4 which processes requested information, a communication equipment 1 which communicates with an external computer, a display device 3 on which windows, graphic elements, characters, etc., are displayed, an input device 2 like a keyboard, a tablet, or a mouse with which data of characters, numerical values, positions, etc., are inputted, a main storage device 5 where a program being in course of execution like an operating system or a window system is stored, and a secondary storage device 6 where programs, text data, etc., are stored, and the secondary storage device 6 is provided with a data analysis module 10 which automatically statistically analyzes electronic information delivered and stored by electronic mails and electronic news to automatically generate topic information in each field.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic information receiving apparatus and an electronic information processing method for processing electronic information distributed by electronic mail, electronic news, and the like.

[0002]

2. Description of the Related Art In recent years, with the spread of the Internet, electronic mail and electronic news distribution services have become popular. Among them, an information providing service called a mailing list (mail list) is enormous in number and types, and news is distributed by e-mail every day if registered in a mailing list of interest or interest. Such distribution and use of electronic information via the Internet is expected to become increasingly widespread in any field.

In a conventional electronic information receiving apparatus, a display device such as a mail system or a news reader displays a symbol for identifying a distributor, a title of a message, a date, and the like in a list, and a reader selects the information. And read the contents.

FIG. 12 is a block diagram showing a conventional electronic information receiving apparatus.

In FIG. 12, reference numeral 61 denotes a communication device such as a modem for performing communication, 62 denotes an input device such as a keyboard and a mouse, 63 denotes a display device for displaying received electronic information, and 64 denotes a central processing unit (CPU). , 65 is a main storage device for temporarily storing the program or value being processed, 66 is a processing module for displaying, editing and transmitting mail information and a secondary storage device for storing and storing mail data. Yes, the secondary storage device 66 includes a mail display module 67 for displaying the received mail,
An e-mail transmission module 68 for performing e-mail transmission;
A document editing module 69 for editing a mail document.

[0006] With regard to the electronic information receiving apparatus configured as described above, the mail display module 67 of the secondary storage device 66 is used.
Will be described with reference to FIG. FIG. 13 shows FIG.
9 is a flowchart showing processing by the mail display module 67 of the electronic information receiving device of FIG.

First, an inquiry is made as to whether a mail has arrived via the communication device 61 (S61). If it is determined that the mail has arrived, the received mail data is stored in the secondary storage device 6.
6 (S62). If it is determined that it has not arrived, the process proceeds to step 63. In step 63, the mail data currently stored in the secondary storage device 66 is read into the main storage device 65 and displayed in a list form. This list does not show the body of the message,
The title is displayed. The mail data includes the contents of the mail, the title of the mail, the mail address of the sender, the transmission time, and the like. Next, a user selection mode is entered (S64). In this selection mode, the user selects the information to be read or the end from the mail list displayed on the screen using the index number or the mouse.
When an index number other than the end is input, the body of the mail data indicated by the index (mail document) is displayed on the screen. When the user has finished reading the body of the mail, the process returns to step 63 to refer to another body of the mail. If the end button is selected in step 64, the processing of the mail display module ends.

Next, fuzzy clustering, which is an algorithm for classifying articles based on certain feature amounts, will be described. Fuzzy clustering is a method of dividing a dispersed data group into several groups.
At this time, the data in the same group has similar properties, and the data is divided so that the average properties between the groups are as different as possible. Now, N data are represented by x1, x2, x3,.
.., Xn, and when this data set is X, the following equation (1) is obtained.

[0009]

(Equation 1)

Consider dividing this into c clusters {2 ≦ c <n}. Now, let ukj be the likelihood that xj belongs to the k-th cluster. In hard clustering, ukj is a binary value of 0 or 1, but in fuzzy clustering, ukj assumes an arbitrary value from 0 to 1.
(Equation 2) indicates hard clustering, and (Equation 3) indicates fuzzy clustering.

[0011]

(Equation 2)

[0012]

(Equation 3)

However, in each case, ukj is (Equation 4)
become that way.

[0014]

(Equation 4)

Fuzzy clustering recognizes that one data belongs to more than one cluster. But,
The sum of the degrees of membership is 1, and u is the weight of membership to each cluster. There is an algorithm for obtaining such u. This algorithm is proposed by Dunn, C. Fuzzy C-m generalized by Bezdek
There is an eans method (C means method), which obtains u using a fuzzy set.

Now, a set of a c × n matrix U (partition matrix) having ukj as an element satisfying (Equation 3) is written as Mfc.
On the other hand, Rd in (Equation 1) is called a feature space, and xk = (x
k1, xk2,..., xkd) are called feature vectors. The clustering is to connect the data set X and the division matrix U. The concatenated one is written as Uk. In the fuzzy clustering algorithm, in order to find the optimum Uk, the sum of the square errors in the next group is used as an objective function and treated as an energy minimization problem. This is (Equation 5)
Shown in

[0017]

(Equation 5)

Here, xj is d-dimensional measurement data, and v
k is the center of the k-th cluster and is d-dimensional. {*} Is an arbitrary norm and represents the similarity between the measurement data and the cluster center. Ukj and vk that minimize the energy function of (Equation 5) satisfy the following equations (Equation 6) and (Equation 7) for m> 1.

[0019]

(Equation 6)

[0020]

(Equation 7)

The ukj that minimizes (Equation 5) can be obtained by repeating the following writing procedure (procedure 1 to procedure 4).

(Procedure 1) Assuming m and the number of clusters c,
The norm is appropriately defined by (Equation 6). Also, an initial value U (0) of U is appropriately given. U (0) is randomly selected regardless of ukj.

(Procedure 2) A cluster center vk (0) is calculated using U (0) and (Equation 7).

(Procedure 3) U (1) is obtained using vk (0) and (Equation 6).

(Procedure 4) An appropriate norm and threshold value ε are defined, and it is determined whether or not {U (p) −U (p−1)} ≦ ε.

Steps 1 to 4 are defined as ‖U (p) -U (p-
1) Repeat until ‖ ≦ ε.

The above is an explanation of the conventional technique of the conventional fuzzy clustering.

[0028]

However, in the conventional electronic information receiving apparatus, while looking at titles and the like from a list of e-mails and selecting data of interest one by one with respect to the distributed e-mail data, If you have to check the content, and if you do not subscribe to multiple email lists or browse for several days, you will receive a huge number of unread emails, and in order to grasp the entire distribution information, It takes time and effort, and the number of unread mails increases in the end, accumulates in the secondary storage device, and has a problem that it is necessary to collectively search when necessary. This is because there is no limit on the number of mail lists registered, so the type of mail list increases as the user's interest increases,
Also, this is because the number of distribution mails increases or decreases according to the distributor's intention.

In the electronic information receiving apparatus and the electronic information processing method, it is required to automatically analyze the contents of the stored mail list data and to inform the user of only data considered to be a topic.

According to the present invention, an electronic information receiving apparatus capable of automatically analyzing the contents of accumulated mail list data and notifying a user of only data considered to be a topic, and the contents of accumulated mail list data An object of the present invention is to provide an electronic information processing method for automatically performing analysis and notifying a user of only data considered to be a topic.

[0031]

According to the present invention, there is provided an electronic information receiving apparatus comprising: a central processing unit for processing requested information; a communication unit for communicating with an external computer; A display device that displays graphic elements, characters, etc., an input device such as a keyboard, tablet, and mouse that inputs data such as characters, numerical values, and positions, and a main memory for storing running programs such as an operating system and a window system Device, a program, an electronic information receiving device having a secondary storage device for storing text data and the like, the secondary storage device,
The system includes a data analysis module that automatically performs statistical analysis on electronic information distributed by e-mail or electronic news and stored in a built-in memory to automatically generate topic information for each field.

Thus, an electronic information receiving apparatus capable of automatically analyzing the contents of the stored mail list data and notifying the user of only data considered to be a topic can be obtained.

In order to solve the above problems, an electronic information processing method according to the present invention comprises a central processing unit for processing requested information, a communication device for communicating with an external computer, a window, graphic elements, characters, and the like. A display device to be displayed, an input device such as a keyboard, tablet, and mouse for inputting data such as characters, numerical values, and positions; a main storage device for storing a running program such as an operating system and a window system; a program; and text. An electronic information processing method in an electronic information receiving device having a secondary storage device for storing data and the like, wherein the electronic information is automatically distributed to electronic information distributed by electronic mail or electronic news and stored in a built-in memory. A configuration with a data analysis step that automatically generates topic information for each field by performing statistical analysis Eteiru.

As a result, an electronic information processing method for automatically analyzing the contents of the stored mail list data and notifying the user of only data considered to be a topic is obtained.

[0035]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An electronic information receiving apparatus according to a first aspect of the present invention comprises a central processing unit for processing requested information, a communication unit for communicating with an external computer, a window, a graphic element, A display device that displays characters and the like, an input device such as a keyboard, a tablet, and a mouse for inputting data such as characters, numerical values, and positions; a main storage device for storing a running program such as an operating system and a window system; An electronic information receiving device having a secondary storage device for storing programs, text data, and the like, wherein the secondary storage device receives electronic information distributed by e-mail or electronic news and stored in a built-in memory. Has a data analysis module that automatically performs statistical analysis and automatically generates topic information for each field. It is.

With this configuration, there is an effect that topic information is automatically determined.

According to a second aspect of the present invention, in the electronic information receiving apparatus according to the first aspect, the data analysis module performs a morphological analysis on the mail document as the stored electronic information. Module, a histogram generation module that performs histogram analysis on the result of the morphological analysis in the morphological analysis module, and a topic keyword candidate that is generated based on the result of the histogram analysis in the histogram generation module. Generate a series histogram,
A topic keyword determination module that determines topic keywords by executing topic keyword fuzzy inference, a clustering module that classifies all stored mail documents based on the topic keywords determined by the topic keyword determination module, and a classification result obtained by the clustering module And a classification result display module for displaying.

With this configuration, all stored mail documents are classified based on the determined topic keyword, the classification result is displayed, and the user is notified of the topic information in detail.

The electronic information processing method according to claim 3, wherein the central processing unit processes the requested information, the communication device communicates with an external computer, and the display device displays windows, graphic elements, characters, and the like. And input devices such as a keyboard, tablet, and mouse for inputting data such as characters, numerical values, and positions; a main storage device for storing running programs such as an operating system and a window system; and programs and text data. An electronic information processing method in an electronic information receiving apparatus having a secondary storage device for performing statistical analysis automatically on electronic information distributed by electronic mail or electronic news and stored in a built-in memory. And a data analysis step for automatically generating topic information for each field.

With this configuration, there is an effect that topic information is automatically determined.

According to a fourth aspect of the present invention, in the electronic information processing method according to the third aspect, the data analyzing step includes performing a morphological analysis on the mail document as the stored electronic information. A histogram generating step of performing a histogram analysis on the result of the morphological analysis in the morphological analysis step; generating a topic keyword candidate based on the result of the histogram analysis in the histogram generating step; A topic keyword determining step of generating a series histogram and executing topic keyword fuzzy inference to determine a topic keyword; a clustering step of classifying all stored mail documents based on the topic keywords determined in the topic keyword determining step; In which it was decided to have a classification result display step of displaying a result of classification in Staring step.

With this configuration, all stored mail documents are classified based on the determined topic keyword, the classification result is displayed, and the topic information is notified to the user in detail.

Hereinafter, an embodiment of the present invention will be described with reference to FIG.
This will be described with reference to FIG.

(Embodiment 1) FIG. 1 is a block diagram showing an electronic information receiving apparatus according to Embodiment 1 of the present invention.
In FIG. 1, reference numeral 1 denotes a communication device such as a modem for communicating with an external computer (for example, an electronic information transmission device); 2 denotes an input device such as a keyboard, tablet, and mouse for inputting data such as characters, numerical values, and positions; A display device for displaying windows, graphic elements, characters, etc., 4 is a central processing unit (CPU) for processing requested information, 5 is a main storage device for temporarily storing programs and values being processed, and 6 is a reception device. And a secondary storage device for storing a processing module, a table, and the like for the obtained electronic information. The secondary storage device 6 includes a mail display module 7, a mail transmission module 8, a document editing module 9, and a data analysis module 10 for analyzing data obtained by storing received mail. The data analysis module 10 includes a morphological analysis module 11 and
It comprises a histogram generation module 12, a topic keyword determination module 13, a topic keyword display module 14, a clustering module 15, and a classification result display module 16. The topic keyword is a keyword for specifying a topic article that frequently appears intensively during a certain period designated by the user.

The operation of the electronic information receiving apparatus configured as described above will be described with reference to FIGS. FIG. 2 shows a mail display module 7 of the electronic information receiving apparatus of FIG.
FIG. 3 is a flowchart showing the operation of the data analysis module 10, FIG. 4 is a flowchart of creating a histogram, FIG. 5 is an explanatory diagram of time series analysis of topic keywords, and FIG. a) to (e) are explanatory diagrams of topic keyword classification, and FIG. 7 is a function diagram showing a fuzzy membership function for topic keyword determination. Note that FIG.
Then, only step 4 shows the operation of the data analysis module 10, and the other steps show the operation of the mail display module 7.

In FIG. 2, first, an inquiry is made as to whether mail data has arrived via the communication device 1 (S
1). If it is determined that the mail has arrived, the received mail data is stored in the secondary storage device 6 (S2). If it is determined that it has not arrived, the process proceeds to step 3. In step 3, the mail data currently stored in the internal memory (not shown) of the secondary storage device 6 is read into the main storage device 5.
Email data includes email content, email title,
Includes the sender's email address, sending time, etc. The mail data read into the main storage device 5 is displayed in a list format (S3). This list does not display the text (mail document) of the mail data, but displays the sender, transmission time, and title. Next, the data analysis module 10 performs data analysis for the designated period of the user from the accumulated past mail list, and displays the result in a list format (S4). Next, it enters the user selection mode (S5). In this selection mode, the user selects the information to be read or the end from the mail list displayed on the screen using the index number or the mouse.
If an index number other than the end is input, the process proceeds to step 6, where the text of the mail data indicated by the index is displayed on the screen. When the user has finished reading the body of the mail data, the process returns to step 3 to refer to another body of the mail data. If the end is selected in step 5, the mail display module ends.

FIG. 3 shows the data analysis processing in step 4 of FIG. The process of FIG. 3 is for realizing a function of extracting important topic keywords from the contents of the mail list accumulated in the past and enabling the user to efficiently check the mail list.

In FIG. 3, first, the morphological analysis module 11 performs a morphological analysis on all mail documents stored in a memory (built-in memory, not shown) in the secondary storage device 5 (S21). In the morphological analysis, a mail document as shown in (Table 1) is decomposed for each word, and the part of speech of the word is recognized as shown in (Table 2).

[0049]

[Table 1]

[0050]

[Table 2]

Here, only the nouns are extracted and set as candidates for topic keywords described later. After performing the morphological analysis, the histogram generation module 12 performs a histogram analysis (S22). The histogram analysis counts up the number of appearances of the noun words extracted as a result of the morphological analysis, so that it is possible to analyze which noun words appear frequently in the accumulated mail document. . Next, the process proceeds to a topic keyword candidate generation process.
As a result of the histogram analysis, the top 5% words are set as topic keyword candidates (S23). Next, the topic keyword determination module 13 shifts to time series histogram generation (S24). Here, the mail document within the period designated by the user is divided into ten in time series, and a histogram is generated again for each topic keyword. Thereafter, the topic keyword determination module 13 executes a topic keyword fuzzy inference, analyzes a histogram frequent tendency for each topic keyword, and determines an appropriate value as a topic keyword (S25). In step 25, the top 10 keywords having the appropriate values of the topic keywords are determined as topic keywords. In step 26, the clustering module 15 classifies all stored mails using the number of appearances of the topic keywords as a feature amount using clustering logic. . Based on this classification result, the result is displayed to the user via the output device (S27). The topic keyword display module 14 displays topic keywords in a histogram or the like.

Next, the histogram processing executed in steps 22 and 24 will be described with reference to FIG.

In FIG. 4, first, a noun word list as a result of morphological analysis is read one by one from the beginning (S31). Next, when it is determined that the list is not at the end (S32), the process proceeds to step 33, where if the word already exists, the counter of the word is incremented (S34). Set up field and count 1
Is set (S35). This operation is repeated in order, and when the list reaches the end in step 32, the histogram table is rearranged in descending order of count (S3
6), this process ends.

Table 3 shows a sample topic keyword candidate list obtained as a result of the histogram processing.

[0055]

[Table 3]

In (Table 3), keywords and histogram counts are stored in descending order of frequency from the top. FIG. 5 shows an example of the result of the histogram analysis based on the time series in step S24. The time series uniformly divides the designated period into ten, and the frequency of occurrence of words in each divided section is shown as a value on the vertical axis of the graph of FIG.

Next, the topic keyword fuzzy inference process in step 25 will be described with reference to FIGS. This fuzzy inference is for determining the suitability of a topic keyword based on rules described in IF-THEN format. The information obtained up to step 24 is a frequent pattern of topic keyword candidates in each time series, and the judgment is slightly different depending on the weight of all words and the period designated by the user. By controlling this delicate determination using fuzzy inference, it becomes possible to perform control that was difficult with the conventional determination of 1,0.

First, topic keywords are classified into five types of patterns as shown in FIG. Class I is the first half peak type, Class II is the middle peak type, Class III is the second peak type, Class IV is the multiple peak type, and Class V is the other type. Which of the classification patterns the time series histogram shown in FIG. 5 is closer to is represented not by the values of 0 and 1 but by a fuzzy number described later and described in a rule.

Table 4 shows examples of fuzzy rules for determining topic keywords.

[0060]

[Table 4]

In the example of (Table 4), the conditions such as the degree of conformity to the classification pattern, the appearance frequency, and the designated period are described in the condition part of the IF by a variable given by a fuzzy number.
In the EN part, the adaptation for calculating the appropriateness value as the trend keyword is expressed as a fuzzy number. The calculation uses the MAX-MIN calculation method.

Next, a method for quantifying the number of fuzzies using the fuzzy membership function for topic keyword determination in FIG. 7 will be described. FIG. 10 shows a VT (Ver
yTrue (1)), RT (Rather True
(2)), PT (Prob-ally True
(3)), VF (Very False (4)), RT
(Rather False (5)), PT (Prob
ably False (6)), UN (Unknown
(7)) shows seven types of fuzzy numbers. The fuzzy number is given by a function called the membership function,
These are indicated by (1) to (7) in the graph in FIG. In order to quantify these, a level value called α cut is used. In the example in the figure, by setting α to 0.75, VF = [0, 0.1] and VT = [0 .9,1.
0] can be obtained.

As described above, the result of the topic keyword determination is obtained in step 25, and the top 10 keywords having the highest suitability values are identified as topic keywords.
After the result display in step 27, the user is notified of the analysis result.

FIG. 8 is a display diagram showing a display example of a topic keyword. In this display example, the target period of the analyzed mail document is displayed, the topic keyword and its suitability value are displayed as importance, and as a result of clustering, the one with the highest degree of belonging to the classification to which it belongs is brought to the front. A news ID icon is displayed.

FIG. 9 is an explanatory diagram of article classification. Each article 1
..The relevance of the keywords contained in n
Perform fuzzy clustering. As a result, each class (cluster) and the degree of belonging of each article to the class can be obtained.

FIG. 10 is a flowchart showing the processing flow of fuzzy clustering. In FIG. 10, first, a keyword set is determined, the number of keywords at this time is set to n, and each keyword is expressed as X = ｛x1, x2, x
3,..., Xn} (S41). Next, the number of classifications k is given appropriately, and the value of the energy function value U (0) for initialization is given (S42). Then, the average vector V of the initial classification U is obtained, and U (0) is changed to U (1) (S43).
A convergence value ε is set, and A = | U (k + 1) −U (k + 2)
Is calculated (S44). This value is | U (k + 1) -U
It is determined whether (k + 2) | ≦ ε is satisfied (S45),
If it is determined that the condition is not satisfied, the process returns to step 42, and this process is repeated. Also, | U (k + 1) -U (k + 2)
If it is determined that | ≦ ε is satisfied, the process ends. The value of k at the end time is the final number of classifications, and the average vector V is the center of the classification. In addition, each value of the matrix vector Unk is a degree of belonging to the classification of each article.

FIG. 11 is a display diagram showing a display example after clustering. On the display screen, a keyword that is a classification central feature amount is displayed for each classification, and a set of articles belonging to the classification is displayed as a list in descending order of the degree of belonging. Articles can be referred to by selecting each article with a pointing device such as a mouse.

As described above, according to the present embodiment, the morphological analysis module 11 that performs morphological analysis on the mail document as the stored electronic information, and the histogram generation that performs histogram analysis on the result of the morphological analysis A module 12 for generating a topic keyword candidate based on the result of the histogram analysis, generating a time-series histogram for the topic keyword candidate, and executing a topic keyword fuzzy inference to determine a topic keyword; A clustering module 15 for classifying all stored mail documents based on the determined topic keyword, and a classification result display module 16 for displaying the classification result in the clustering module. It is possible to display the classification result to classify all the stored mail document, the user can know the topic information easier, it is possible to obtain effective and efficient beneficial topical information.

[0069]

As described above, according to the electronic information receiving apparatus of the first aspect of the present invention, a central processing unit for processing requested information, a communication apparatus for communicating with an external computer, a window, A display device for displaying graphics, graphic elements, characters, etc., a keyboard, tablet, mouse, etc. for inputting data such as characters, numerical values, positions, etc., and a main unit for storing running programs, such as an operating system and a window system. A storage device, a program,
An electronic information receiving device having a secondary storage device for storing text data and the like, wherein the secondary storage device automatically distributes electronic information distributed by electronic mail or electronic news and stored in a built-in memory. By having a data analysis module that automatically performs statistical analysis and automatically generates topic information for each field, topic information for each field is automatically generated, so that topic information can be automatically determined. The advantageous effect that it can be obtained is obtained.

According to the electronic information receiving apparatus of the second aspect, in the electronic information receiving apparatus of the first aspect, the data analysis module performs a morphological analysis on the mail document as the stored electronic information. A morphological analysis module, a histogram generation module that performs a histogram analysis on the result of the morphological analysis in the morphological analysis module, and a topic keyword candidate based on the result of the histogram analysis in the histogram generation module. Topic keyword determination module that generates a time series histogram and performs topic keyword fuzzy inference to determine topic keywords, and a cluster that classifies all stored mail documents based on the topic keywords determined by the topic keyword determination module And a classification result display module that displays the classification result in the clustering module, the classification result can be displayed by classifying all the stored mail documents based on the determined topic keyword. And the topic information can be easily known, and useful topic information can be obtained effectively and efficiently.

According to the electronic information processing method of the present invention, a central processing unit for processing requested information, a communication device for communicating with an external computer, and windows, graphic elements, characters, and the like are displayed. Display device, characters, numbers,
Input devices such as a keyboard, tablet, and mouse for inputting data such as positions, a main storage device for storing running programs such as an operating system and a window system, and a secondary storage for storing programs, text data, and the like. An electronic information processing method in an electronic information receiving apparatus having a device, wherein statistical information is automatically analyzed for electronic information distributed by electronic mail or electronic news and stored in a built-in memory, and topic information for each field is provided. Is automatically generated, the topic information for each field is automatically generated, so that an advantageous effect that the topic information can be automatically determined is obtained.

According to the electronic information processing method of the fourth aspect, in the electronic information processing method of the third aspect, the data analysis step performs a morphological analysis on the mail document as the stored electronic information. A morphological analysis step;
A histogram generation step of performing a histogram analysis on the result of the morphological analysis in the morphological analysis step, and generating a topic keyword candidate based on the result of the histogram analysis in the histogram generation step, and generating a time-series histogram for the topic keyword candidate A topic keyword determining step of generating and executing a topic keyword fuzzy inference to determine a topic keyword; a clustering step of classifying all stored mail documents based on the topic keyword determined in the topic keyword determining step; and a clustering step. By having the classification result display step of displaying the classification result, all the stored mail documents can be classified based on the determined topic keywords and the classification result can be displayed. The click information can be easily known,
The advantageous effect that useful topic information can be obtained effectively and efficiently is obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an electronic information receiving apparatus according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing operations of a mail display module and a data analysis module of the electronic information receiving device of FIG.

FIG. 3 is a flowchart showing the operation of the data analysis module;

FIG. 4 is a flowchart for creating a histogram.

FIG. 5 is an explanatory diagram of a time series analysis of topic keywords.

FIG. 6 (a) Illustration of topic keyword classification (b) Illustration of topic keyword classification (c) Illustration of topic keyword classification (d) Illustration of topic keyword classification (e)

FIG. 7 is a function diagram showing a fuzzy membership function for topic keyword determination.

FIG. 8 is a display diagram showing a display example of a topic keyword.

FIG. 9 is an explanatory diagram of article classification.

FIG. 10 is a flowchart showing a processing flow of fuzzy clustering.

FIG. 11 is a display diagram showing a display example after clustering.

FIG. 12 is a block diagram showing a conventional electronic information receiving apparatus.

FIG. 13 is a flowchart showing processing by the mail display module of the electronic information receiving device in FIG. 12;

[Explanation of symbols]

 Reference Signs List 1 communication device 2 input device 3 display device 4 central processing unit (CPU) 5 main storage device 6 secondary storage device 7 mail display module 8 mail transmission module 9 document editing module 10 data analysis module 11 morphological analysis module 12 histogram generation module 13 Topic keyword judgment module 14 Topic keyword display module 15 Clustering module 16 Classification result display module

Continued on the front page F-term (reference) 5B075 NK32 NR12 PP02 PP03 PQ02 UU24 5B089 GA21 JA31 KA04 KC32 KC59 LB14 5K030 GA17 HA06 HB00 HB19 KA01 KA02 LE16 9A001 BB02 BB03 BB04 CC02 FF01 FF03 JJ03 KK03 JJ FF03 GG07

Claims (4)

[Claims] 1. A central processing unit for processing requested information, a communication device for communicating with an external computer, a display device for displaying windows, graphic elements, characters, etc., and data for characters, numerical values, positions, etc. It has an input device such as a keyboard, tablet, and mouse for input, a main storage device for storing a running program such as an operating system and a window system, and a secondary storage device for storing a program, text data, and the like. An electronic information receiving device, wherein the secondary storage device automatically performs statistical analysis on electronic information distributed by electronic mail or electronic news and stored in a built-in memory to automatically obtain topic information for each field. An electronic information receiving apparatus, comprising: a data analysis module that generates the data. 2. The data analysis module according to claim 1, further comprising: a morphological analysis module configured to perform a morphological analysis on the mail document as stored electronic information; and a histogram configured to perform a histogram analysis on a result of the morphological analysis performed by the morphological analysis module. A generating module, generating topic keyword candidates based on the result of the histogram analysis in the histogram generating module, generating a time-series histogram for the topic keyword candidates, executing topic keyword fuzzy inference to determine topic keywords A topic keyword determination module, a clustering module for classifying all stored mail documents based on the topic keywords determined by the topic keyword determination module, and a classification result by the clustering module. Electronic information receiving apparatus according to claim 1, characterized in that it comprises a classification result display module. 3. A central processing unit for processing requested information, a communication device for communicating with an external computer, a display device for displaying windows, graphic elements, characters, etc., and data for characters, numerical values, positions, etc. It has an input device such as a keyboard, tablet, and mouse for input, a main storage device for storing a running program such as an operating system and a window system, and a secondary storage device for storing a program, text data, and the like. An electronic information processing method in an electronic information receiving apparatus, wherein statistical analysis is automatically performed on electronic information distributed by electronic mail or electronic news and stored in a built-in memory to automatically obtain topic information for each field. An electronic information processing method having a data analysis step of generating. 4. The data analyzing step includes a morphological analysis step of performing a morphological analysis on the mail document as accumulated electronic information, and a histogram performing a histogram analysis on a result of the morphological analysis in the morphological analyzing step. Generating a topic keyword candidate based on the result of the histogram analysis in the generating step and the histogram generating step, generating a time-series histogram for the topic keyword candidate, and executing topic keyword fuzzy inference to determine the topic keyword Topic keyword determination step, a clustering step of classifying all stored mail documents based on the topic keyword determined in the topic keyword determination step, and a classification result display for displaying a classification result in the clustering step Electronic information processing method according to claim 3, characterized in that it comprises a step.