JPH11296522A - System and method for deciding attribute of information and record medium where attribute deciding program is recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provided the system and method for information attribute decision making which decide the attribute of information more precisely and faster and the record medium where the attribute deciding program for information is recorded. SOLUTION: A word representing the attribute of information included in information consisting of a text or elements which can be converted into the text is extracted from the information and this extracted work is given a numeral according to specific algorithm; and obtained numerals are combined according to other specific algorithm to obtain a different numeral, with which the attribute of the information is quantitatively decided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a text or an element which can be converted into a text, and quantitatively determines the attribute (eg, importance or urgency) of the information to be stored or transmitted / received. The present invention relates to an information attribute determination system and method, and a recording medium on which an information attribute determination program is recorded, by which information attributes can be easily determined by representing the information attributes so that a user of the information can efficiently process the information. Things.

[0002]

2. Description of the Related Art Conventionally, for example, WWW, e-mail, and NEW have been used so that information users can process information efficiently.
In information media such as S, mailing list, telephone, facsimile, and television, and services such as information search (directory service, page search service, etc.) and information distribution service (e-mail newspaper), the user himself / herself can collect information according to his / her taste. There is a method of performing selection and rearrangement, or performing selection and rearrangement by registering a user's age, gender, and keywords (words representing a target area such as “DOS / V” and “Italian food”). . For example, you can tune to a TV channel, or register a keyword and fax the information that matches the keyword.
It is a method of transmitting by. In addition, the presentation order is determined based on the recent user's tendency (that is, fashion), and the presentation order of information to the user is automatically determined based on the tendency of other people who have similar items indicating the characteristics of the user. Or there is a way.

[0003]

In recent years, enormous information has been accumulated in information media and transmitted and received. In order for users to obtain and process information efficiently, it is important to select information that is useful (eg, important, urgent, interesting) to the user. However, when determining attributes of information (for example, importance or urgency) with higher accuracy, it is necessary to understand the details of the information.

In general, in order to completely grasp the attributes of text information, it is necessary to analyze the text information using a natural language processing technique. However, it is impossible to completely automatically analyze information attributes using natural language processing technology, or even if it is possible, it requires enormous time and is not practical.

[0005] Usually, in order to obtain the attributes necessary for selecting information meaningful to the user, the entire text information is not necessary, and in many cases, some expressions included in the information are necessary. Represents attributes.

Accordingly, an object of the present invention is to extract expressions representing a certain attribute included in information, process each expression by an appropriate means, and use the processing result for selecting information. Accordingly, it is an object of the present invention to provide an information attribute determination system and method for determining an information attribute with high accuracy and high speed, and a recording medium on which an information attribute determination program is recorded.

[0007]

In order to achieve the above object, an information attribute determination system according to the present invention is directed to a method for extracting a word representing an attribute of information included in text or information composed of elements that can be converted to text. Word extracting means for extracting from the information, first numerical value assigning means for assigning a numerical value to the words extracted by the word extracting means in accordance with a predetermined algorithm, and a numerical value obtained by the first numerical value assigning means, A second numerical value providing means for obtaining a numerical value for each combination according to a predetermined algorithm; and a determining means for quantitatively determining an attribute of the information based on the numerical value obtained by the second numerical value providing means. It is a feature.

[0008] The information attribute determining method according to the present invention includes a text data reference step of extracting, from the information, a word representing an attribute of information contained in information composed of text or elements that can be converted to text. An expression table reference step in which the words extracted in the data reference step are referred to as an expression table to determine whether an expression in the expression table exists, and a word determined in the expression table reference step is referred to an expression database. Then, referring to the expression database reference step for pricing based on the expression database value, and the numerical value obtained in the expression database reference step as a calculation formula database,
A calculation formula reference step of substituting the value of each expression into a calculation formula to calculate the value of the information, and a determination step of quantitatively determining the attribute of the information based on the numerical value obtained in the calculation formula reference step. It is characterized by having.

The recording medium storing the information attribute judging program of the present invention allows a computer to extract, from the information, a word representing the attribute of the information contained in the information composed of text or elements that can be converted to text. A first procedure, a second procedure for assigning a value for each expression to the words extracted in the first procedure, and substituting the value for each expression obtained in the second procedure into a calculation formula to obtain the information And a fourth procedure for quantitatively determining the attribute of the information based on the numerical value obtained in the third procedure. An information attribute determination program for executing the information is recorded. It is characterized by the following.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings. 1 and 2 are flowcharts showing an embodiment of the present invention. The information attribute determination system of the present invention extracts words representing attributes of information included in text or information which can be converted to text from the information and converts the words into the words shown in FIG. Calculating the numerical values according to the algorithm, and further combining the obtained numerical values according to the algorithm of FIG. 2 to obtain different numerical values, and the attribute of the information (for example, importance, (Urgency) is calculated and the attribute of the information is quantitatively determined.

That is, as shown in FIG. 1, first, in a text data reference step, text data is referred to, and a word representing an attribute of information contained in text or information composed of elements that can be converted to text is written in the text data reference step. Extract from information. Next, in the expression table reference step, the words extracted in the text data reference step are referred to as an expression table, and it is determined whether or not an expression in the expression table exists. Next, when there is an expression in the expression table, in the expression database reference step, the words determined in the expression table reference step are referred to as the expression database, and a price is given based on the expression database value. Get the value for each expression. If the expression in the expression table does not exist, a price is assigned with reference to a default value when there is no expression in the expression database, and a value for each expression in the information is obtained. In the expression table, types of expressions, types of expressions,
Are stored in the expression database, [expression type, expression, value, ...] and [expression type, if none, value,
…]].

Next, as shown in FIG. 2, in an expression table reference step, an expression corresponding to a numerical value obtained according to the algorithm of FIG. 1 is extracted. Next, in the calculation formula reference step, the calculation formula in the target area of the information and its coefficient database are referred to, in order to obtain the value for each expression in the information, the value for each expression is substituted into the calculation formula, and the value of the information is calculated. Is calculated to obtain a determination value of the information for quantitatively determining the attribute of the information.

It is to be noted that a computer is provided with a first method for extracting words representing attributes of information contained in information composed of text or elements which can be converted into text from the information.
, A second procedure for assigning a value for each expression to the words extracted in the first procedure, and substituting the value for each expression obtained in the second procedure into a calculation formula to obtain the value of the information Recording medium for recording an information attribute determination program for executing a third procedure for calculating the information, and a fourth procedure for quantitatively determining the attribute of the information based on the numerical value obtained in the third procedure. May be configured.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As an embodiment of the present invention, an example of calculation of importance and urgency of information for a business email for user support will be described below. As shown in the example of FIG.
Please let us know the schedule of the telephone construction tomorrow.
When an e-mail of "XX company xx" is received, the expression "relationship between receiver and sender", "keyword" representing an attribute from an expression table as shown in FIG. ”,“ Date ”, and“ Tension ”are extracted. 9 to 1 for each description.
2 is searched from each expression database, and a calculation formula is applied. FIG. 9 shows an example of a value based on a keyword.
0 is an example of a value based on a tension expression, FIG. 11 is an example of a value for a relationship between a receiver and a sender, and FIG. 12 is an example of a value based on a period length.

That is, the relationship between the receiver and the sender is α = 3 in “affiliated company”, and the keyword is “telephone work”.
And β = 5. Therefore, (degree of interest) = a × α + b
× β = 8. The coefficients a and b are set to 1.

The length of the period from the date of arrival to the date of arrival is:
Γ = 1 in “the day after tomorrow”, and the urgency expression is δ in “urgent”
= 5. Therefore, (tensity) = c × γ + d × δ
= 9. The coefficients c and d are 1.

From the above, (priority) ² = (interest) ² +
(Tension) ² = 8 ² + 9 ² = 64 + 81 = 145 In addition, as shown in the example of FIG.
It is an introduction of the product. Look at the XX homepage. Until tomorrow. Automatically received for each e-mail, the expressions "relationship between recipient and sender", "keyword", "date", and "relationship between recipient and sender" representing an attribute from an expression table as shown in FIG. The description of "tense expression" is extracted. For each description, a value for it is retrieved from each of the expression databases in FIGS.
Apply the formula.

That is, the relationship between the receiver and the sender is α = 2 without, and the keyword is β = 3 without.
Therefore, (degree of interest) = a × α + b × β = 5.
The coefficients a and b are set to 1.

The length of the period from the due date to the arrival date is γ = 1 without any, and the urgency expression is δ = 5 in “until tomorrow”.
It is. Therefore, (tensity) = c × γ + d × δ = 6
Becomes The coefficients c and d are 1.

From the above, (priority level) ² = (interest level) ² +
(Tension) ² = 5 ² +6 ² = 25 + 36 = 61 FIG. 7 is an example of a formula for calculating the priority, and an example of calculating the importance and urgency of information is the formula in FIG. Here, the degree of interest in the formula is the degree of interest of the user in the information, and is calculated by a relationship between the receiver and the sender and a keyword indicating the interest of the user. The degree of urgency is the degree of urgency of the information for the user, and is calculated from the date and the expression of the urgency. Based on the degree of interest and the degree of urgency, the degree of importance and urgency of the information is calculated as the priority. In addition, when there are a plurality of descriptions corresponding to the items of “relationship between receiver and sender”, “keyword”, “date”, and “tense expression”, the one with the larger value is selected. The degree of interest in the calculation formula of FIG. 7 is calculated from the values obtained by “the relationship between the receiver and the sender” and “keyword”. Also,
The degree of urgency in the calculation formula of FIG. 7 is calculated from the values of “date” and “tense expression”. The priority is calculated based on the degree of interest and the degree of urgency. This is the priority of one e-mail.

In this example, high values are set and calculated in all items of “relationship between receiver and sender”, “keyword”, “date”, and “tense expression” in FIG. Only the expression "high" is set and calculated. Therefore, the order of importance and urgency for the user is the e-mail in FIG. 3 and the e-mail in FIG. Here, the pricing to words and the coefficients (a, b, c, d) in the formula are determined by the area to which the formula is applied.

Next, FIG. 5 and FIG. 6 can be considered as an electronic mail system using the determination of the importance and urgency of this information. FIG. 5 shows an information distribution system constructed on an electronic mail server. As a server function,
By calculating the importance and urgency of the e-mail that arrives, the terminal performance of mobile terminals and mobile phones etc.
In an environment with low communication performance, important and urgent mails can be selectively retrieved, and there is no need to retrieve all arriving e-mails.

That is, first, e-mails are sequentially extracted from a plurality of e-mails, keyword data is extracted from the keyword database for the extracted e-mails, and priced based on the keyword data. Also,
With respect to the extracted electronic mail, the organization and the human relationship data are extracted from the sender's organization and the human relationship database, and are priced based on the sender's organization and the human relationship data. Further, for the retrieved e-mail, tension expression data is retrieved from the tension expression database and priced based on the tension expression data. Further, with respect to the extracted e-mail, period data is extracted from the date expression database, and is priced based on the period data. The degree of interest is calculated from the pricing based on the keyword data and the pricing based on the organization and human relationship data of the sender. In addition, the degree of urgency is calculated from the valuation based on the urgency expression data and the valuation based on the period data. A priority is calculated from the degree of interest and the degree of urgency, and a plurality of e-mails are prioritized. When a user requests delivery of (xx) e-mails, only one of the e-mails (xx only)
Send in priority order.

FIG. 6 shows the function of the electronic mail tool of the user terminal. If the importance and urgency of an e-mail are calculated from time to time, it is possible to select necessary information from a large number of mails.

That is, first, electronic mails are sequentially extracted from the server, keyword data is extracted from the keyword database for the extracted electronic mails, and the electronic mails are priced based on the keyword data. Further, with respect to the extracted electronic mail, the organization and the human relation data are extracted from the organization and the human relation database of the sender, and the electronic mail is priced based on the organization and the human relation data of the sender. Further, for the retrieved e-mail, tension expression data is retrieved from the tension expression database and priced based on the tension expression data. Further, with respect to the extracted e-mail, period data is extracted from the date expression database, and is priced based on the period data. The degree of interest is calculated from the pricing based on the keyword data and the pricing based on the organization and human relationship data of the sender. In addition, the degree of urgency is calculated from the valuation based on the urgency expression data and the valuation based on the period data. A priority is calculated from the degree of interest and the degree of urgency, and a plurality of e-mails are prioritized and displayed in order of priority among the plurality of e-mails.

If the electronic mail in FIG. 5 is a voice message, it indicates a voice message (answering machine) system stored in the network. It is possible to select and hear a contact from an important dealer or an urgent inquiry. This is considered to be a useful function because the method of listing stored voice messages is poor on the telephone.

[0027]

As described above, according to the present invention, it is possible to determine the attributes (eg, importance, urgency) of information to be stored or transmitted / received at high speed. Therefore, it is possible to avoid that only necessary information is obtained and that necessary information is buried in unnecessary information and cannot be found, and the user can obtain and process information efficiently, which is beneficial.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a part of an embodiment of the present invention.

FIG. 2 is a flowchart showing a part of an embodiment of the present invention.

FIG. 3 is an explanatory diagram showing one embodiment of the present invention.

FIG. 4 is an explanatory view showing another embodiment of the present invention.

FIG. 5 is a flowchart showing an information distribution system constructed on an e-mail server according to the present invention.

FIG. 6 is a flowchart showing a functional system of an electronic mail tool of a user terminal according to the present invention.

FIG. 7 is an explanatory diagram showing an example of a calculation formula for obtaining a priority according to the present invention.

FIG. 8 is an explanatory diagram showing an example of an expression representing a certain attribute according to the present invention.

FIG. 9 is an explanatory diagram showing an example of a value based on a keyword according to the present invention.

FIG. 10 is an explanatory diagram showing an example of a value by a tension expression according to the present invention.

FIG. 11 is an explanatory diagram showing an example of a value for a relationship between a receiver and a sender according to the present invention.

FIG. 12 is an explanatory diagram showing an example of a value according to a period length according to the present invention.

[Explanation of symbols]

α Value based on the relationship between the receiver and the sender β Value for the keyword γ Value based on the length of the period from the date of arrival to the date of arrival δ Value for the urgency expression a, b, c, d Coefficient in the target area where the formula is to be applied

Claims (3)

[Claims] 1. A word extracting means for extracting a word representing an attribute of information included in information composed of text or an element which can be converted to text from the information; A first numerical value assigning means for assigning a numerical value according to an algorithm, and a second numerical value obtaining a different numerical value by combining the numerical value obtained by the first numerical value assigning means according to another predetermined algorithm.
An information attribute determining system comprising: a numerical value assigning means; and a determining means for quantitatively determining an attribute of the information based on the numerical value obtained by the second numerical value assigning means. 2. A text data reference step for extracting a word representing an attribute of information contained in information composed of text or an element which can be converted to text from the information, and a word extracted in the text data reference step. An expression table reference step for referring to an expression table to determine whether or not an expression in the expression table exists. A word determined in the expression table reference step is referred to an expression database, and the expression database value is assigned. An expression database reference step, and a numerical expression obtained in the expression database reference step is referred to as a calculation expression database, and a value of each expression is substituted into a calculation expression to calculate a value of the information; The attribute of the information is quantitatively determined by the numerical value obtained in this calculation formula reference step. Attribute determination method of information, characterized by comprising a determination step of. 3. A first procedure for extracting, from the information, a word representing an attribute of information included in information composed of text or an element which can be converted to text, the computer comprising: A second procedure for assigning a value for each expression to a word; a third procedure for calculating the value of the information by substituting the value for each expression obtained in the second procedure into a calculation formula; And a fourth step of quantitatively determining the attribute of the information based on the numerical value obtained in the step (a).