JPH06243173A - Personal adaptive type network control method - Google Patents

Abstract

PURPOSE:To execute the processing desired by a user without user's input of indication information. CONSTITUTION:A user action monitor part 16 stores results, which are obtained by discriminating whether the user requires or does not require presented information, in an action data base 17 correspondingly to keywords included in information. An analysis part 18 obtains evaluation values of degrees of importance of keywords from contents of the action data base 17 and writes respective evaluation values in a keyword evaluation table 19. A service start part 20 refers to the keyword evaluation table 19 to instruct an information gathering service execution part 14 to gather the information including a keyword whose evaluation value is high. The information gathering service execution part 14 retrieves information including the indicated keyword from an information data base 21n and presents this information to the user.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a network control method for realizing processing adapted to the behavior of a user.

[0002]

2. Description of the Related Art Conventionally, the problems to be solved by the invention
When a user searches for information in a database via the network, it is necessary to input a keyword to instruct the network what information to search.

With the progress of the information-oriented society, the functions of networks are becoming more sophisticated, and the types of services provided to users and the amount of information provided by each service are increasing. In that case, the amount of instruction information that the user needs to give to the network in order to obtain the desired information increases, not only does it take time to operate, but it becomes difficult for general users to use because the input operation becomes complicated. The problem arises.

[0004] For example, when the information collection service provided by the network is used to retrieve the necessary information from the database, the user needs to input the instruction information for instructing the information to be retrieved. As the scale increases and the amount of information in the database increases, the amount of instruction information that needs to be input becomes more and more complicated.

An object of the present invention is to provide a network control method which allows a user to execute desired processing without inputting instruction information.

[0006]

[Means for Solving the Problems] FIGS. 1 (a) and 1 (b) show the first
FIG. 3 is a principle explanatory diagram of a personal adaptive network control method of the second invention.

In FIG. 1A, in the network control method of the first invention, the behavior of the user of the network is monitored and stored (1), and the behavior of the user is analyzed to extract a rule pattern (2 ), Based on the extracted rule pattern, processing corresponding to each user is executed (3).

In FIG. 1 (b), in the network control method of the second invention, the importance of the keyword contained in the information is determined for each user based on the judgment result of the necessity / unnecessity of each user for the presented information. (4), and the information including the keywords that have been evaluated with high importance is selectively collected and presented to the user (5).

[0009]

In the personal adaptive network control method according to the first aspect of the invention, since the processing corresponding to each user is executed from the rule pattern of the past behavior of the user, the behavior pattern of the user is temporarily stored. After that, even if the user does not give the instruction information to the network, a process matching the user's request is executed.

Further, in the personal adaptive network control method of the second invention, a keyword included in the information which the user has determined necessary in the past with a high probability is evaluated as a highly important keyword, and the keyword is used. When collecting the same type of information or related information, the user should obtain the necessary information without inputting the keyword, because the information retrieved by searching is automatically presented to the user. You can

Further, based on the temporal transition of the user's past judgment result of necessity / unnecessity, there is a tendency that the probability of being included in the information judged necessary by the user is increasing.
Since it is evaluated as a keyword having a high degree of importance and information including the keyword is automatically presented to the user, it is possible to input necessary information without performing any keyword input operation.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 is a system configuration diagram of an embodiment in which the personal adaptive network control method of the present invention is applied to an intelligent network.

This system is composed of a service control node 11 that realizes various communication services, an exchange 12 that constitutes a communication network, and a plurality of user terminal devices 13 that are connected to the exchange 12.

The service control node 11 is composed of an information collection service execution unit 14 and a plurality of information collection control units 15 ₁ to 15 _m provided corresponding to individual users. Each information collection control unit 15 _i includes a user behavior monitoring unit 16, a behavior database 17, an analysis unit 18, a keyword evaluation table 19, and a service activation unit 20.

The information collection service execution unit 14 stores the information including the keyword instructed according to the instruction from the service activation unit 20 in the information database 21 _i (21 _{1 to} 21 _n ).
The obtained information is presented to the user and the user's judgment result for the information is obtained.

The user behavior monitoring unit 16 monitors the information presented to the user by the information service execution unit 14 and the result of the user's decision regarding the information, and the keyword contained in the presented information and the user's decision. The results are associated with each other and stored in the behavior database 17.

FIG. 3 is a diagram showing an example of the configuration of the behavior database 17. Every time the information is presented to the user, the action database 17 writes the determination result made by the user for the information for all the keywords included in the information.

For example, when the information collection service execution unit 14 presents the information 1 including two keywords of "PBX" and "service" to the user and the user judges that the information 1 is unnecessary, The user behavior monitoring unit 16 writes “x” in the relevant keyword in the behavior database 17. At this time, if the corresponding keyword is not registered in the behavior database 17, after the keyword is registered, “x” is written in the keyword.

In addition, "○ × company", "PBX", "TC
The information 2 including the three keywords “SI” is presented to the user, and when the user determines that the information is necessary, “◯” is written in those keywords of the behavior database 17.

Whether the user determines that information is necessary,
The determination as to whether or not it is unnecessary is made based on whether or not the presented information has been downloaded by the user. If the user confirms the presented information but does not download it, it is determined that the information is deemed unnecessary.

The analysis unit 18 obtains the rule pattern of the user's action, that is, the evaluation value of the importance of the keyword from the content of the action database 17, and writes the evaluation value in the keyword evaluation table 19.

FIG. 4 is a diagram showing an example of the configuration of the keyword evaluation table 19. Keyword evaluation table 1
“-9” indicates the importance of each keyword.
The evaluation values in four stages from "+++" are written. A general keyword is registered as a default in the keyword evaluation table 19 as an initial state, and thereafter, an evaluation value is written based on the keyword included in the information presented to the user and the determination result thereof. Going away.

The service activation section 20 refers to the keyword evaluation table 19 provided for each user to instruct the information collection service execution section 14 to collect information including a keyword having a high evaluation value. Upon receiving this instruction, the information collection service execution unit 14 collects information including the instructed keyword from the information database 21n and presents it to the user.

In this case, the information is selectively collected using the keywords included in the information that has been highly necessary by the user in the past in response to the access request from the user. When collecting information, the user can obtain necessary information without giving instruction information to the network.

Next, an evaluation method for determining the evaluation value of the keyword evaluation table 19 in the analysis unit 18 will be described. FIG. 5 is a flow chart for explaining a keyword evaluation method (1) for probabilistically estimating the importance of a keyword from the past judgment result of the user.

First, in step S1 of FIG. 5, “◯” and “x” on the behavior database 17 for a certain keyword.
Add up the number of Next, in step S2, the number of "○" / (the number of "○" + the number of "×") is calculated, and the ratio P of the keyword being "○", that is, the information including the keyword is used. The ratio P which is presented to the person and judged necessary is obtained.

Then, in step S3, the ratio P in which the keyword obtained by the above calculation is "O" is P> 0.3,
It is determined whether 0.1 <P ≦ 0.3, 0.05 ≦ P ≦ 0.1, or P <0.05.

The ratio P in which the keyword is "○" is 0.3
If the value is larger, it means that in the past, the ratio of the user determining that the information including the keyword is necessary is the highest, so the process proceeds to step S4, and the evaluation value of the keyword in the keyword evaluation table 19 is one of four evaluation values. Write the highest evaluation value, "+++".

In step S3, the keyword is "○"
If the ratio P was 0.1 <P ≦ 0.3, it means that the ratio of the user's determination that the information including the keyword was necessary was the second highest in the past, so the process proceeds to step S5. As the evaluation value of the keyword in the advanced keyword evaluation table 19, 2 out of the four evaluation values
Write the highest evaluation value "++".

Further, the ratio P in which the keyword is "○"
Is within the range of 0.05 ≦ P ≦ 0.1, it means that the ratio of the user's judgment that the information including the keyword is necessary is the third highest in the past, so step S6
Then, as the evaluation value of the corresponding keyword in the keyword evaluation table 19, “+”, which is the third highest evaluation value among the four evaluation values, is written.

Further, when the ratio P in which the keyword is "○" is smaller than 0.05, it is judged that the information including the keyword is not necessary for the user, so the process proceeds to step S7 and the keyword evaluation table. As the evaluation value of 19 relevant keywords, “-”, which is the lowest evaluation value among the four evaluation values, is written.

In this evaluation method, the evaluation value of the importance of the keyword is stochastically determined based on the past evaluation behavior of the user with respect to the presented information, and the keyword matching the user's wish is estimated. Therefore, when collecting the same information that the user has recently needed, a keyword input operation becomes unnecessary.

Next, FIG. 6A is a flow chart for explaining a keyword evaluation method (2) for evaluating the importance of the keyword from the transition tendency of the user's past judgment results. First, in step S11 of FIG. 6 (a), the number of “◯” and “×” is aggregated for each keyword in the action database 17 for each period of FIG. 6 (b), and the number of “◯” /
Calculate the value of (the number of "○" + the number of "x").

FIG. 6 (b) is a diagram showing a total period for one keyword, and a ratio P in which the keyword is “◯” is calculated for each period indicated by an arrow for the same keyword.

Next, in step S12, the transition of the ratio P in which the keyword in each period is "O", that is, the gradient k when the ratio is plotted on the horizontal axis and the vertical axis on the ratio P is determined. Then, in step S13, the gradient k is
k> 0.5, 0.3 <k≤0.5, 0.1 ≤k≤0.3, k
Determine if <0.1.

If the gradient k is larger than 0.5 in the determination in step S13, it means that the gradient k showing the temporal transition of the result that the information including the keyword is determined to be necessary by the user is the largest. "+++" which is the highest evaluation value among the evaluation values of four levels as the evaluation value of the keyword in the keyword evaluation table 19
Write.

In step S13, the gradient k is 0.3 <k.
When it belongs to the range of ≦ 0.5, the gradient k indicating the temporal transition of the result that the information including the keyword is determined to be necessary by the user is the second largest, so the process proceeds to step S15 and the keyword evaluation table 19 is searched. As the evaluation value of the keyword, "++", which is the second highest evaluation value among the evaluation values of four levels, is written.

When the gradient k belongs to the range of 0.1 ≤ k ≤ 0.3, the gradient k indicating the temporal transition as a result of the judgment that the information including the keyword is necessary for the user is the third largest. Therefore, the process proceeds to step S16 and the evaluation value of the keyword in the keyword evaluation table 19 is "+" which is the third highest evaluation value among the evaluation values of four levels.
Write.

Further, when the gradient k is smaller than 0.1, it means that the gradient k showing the temporal transition of the judgment result judged to be necessary by the user is the smallest, so the routine proceeds to step S17, and the keyword of the keyword evaluation table 19 is reached. As the evaluation value of, the lowest evaluation value "-" among the four evaluation values is written.

In this case, the importance of the keyword is determined from the past trend of judgment of the user, and the keyword matching the user's wish is estimated based on the importance.
Therefore, when the behavior database 17 and the keyword evaluation table 19 are constructed to some extent, when collecting the information of the same kind as the information judged to be necessary recently or the information related to the information, the user indicates the information to be searched. You can get the necessary information without entering keywords.

According to this evaluation method, a high evaluation value is given to a keyword in which the ratio determined to be necessary by the user tends to increase, and a low evaluation value is given to a keyword which the user does not need much at present. , It is possible to selectively collect only the information including the keyword that the user currently determines to be necessary.

Next, FIG. 7 is a flow chart for explaining a keyword evaluation method (3) which is a combination of the above two evaluation methods. First, in step S21 of FIG. 7, the number of “◯” s and the number of “x” s in the behavior database 17 are aggregated for a certain keyword.

Next, in step S22, the number of "○" /
(The number of “◯” + the number of “×”) is calculated, the evaluation value of the keyword is obtained, and the obtained evaluation value is set as α. Next, in step S23, for the same keyword, the number of “◯” / (the number of “◯” + the number of “x”) is calculated on the behavior database 17 for each period of FIG. 6 (b). In step S24, a gradient k indicating the temporal transition of the calculated values in each period is obtained, and an evaluation value corresponding to the value of the gradient k is set as β.

Then, in step S25, the evaluation value of the keyword is written in the keyword evaluation table 19 based on the larger value of α and β. In the above evaluation method, by determining the importance of the keyword based on both the past judgment result of the user and the temporal transition of the judgment result, the keyword more suited to the user's wish is estimated. be able to.

Next, FIG. 8 is a flow chart for explaining a keyword evaluation method (4) for evaluating the importance of the combination of two keywords. First, step S3 in FIG.
At 1, two keywords A and B for which the importance of the combination is to be evaluated are generated. Next, in step S32, the number of both “A” and “X” of the keywords A and B on the behavior database 17 is totaled.

Further, in step S33, the keywords A and B are both the number of “◯” / (the number of both “+” +
Both calculate the number of "x"). Then, in the next step S34, the calculated value P is P> 0.3, 0.1 <P ≦ 0.
3, 0.05 <P ≦ 0.1, P <0.1 is determined.

When the calculated value is larger than 0.3 in the determination of step S34, the two keywords A and B are both "○".
Is the highest, and in this case, the process proceeds to step S35, and as the evaluation value of the corresponding keyword in the keyword evaluation table 19, the highest evaluation value “+++” among the evaluation values of four levels is written. .

When the calculated value P belongs to the range of 0.1 <P ≦ 0.3, the ratio of both the keywords A and B being “◯” is the second highest, and in this case, the step S
In step 36, “++”, which is the second highest evaluation value among the four evaluation values, is written in the keyword evaluation table 19.

When the calculated value P belongs to the range of 0.05 <P ≦ 0.1, the two keywords A and B are both “◯”.
Is the third highest, and in this case, the process proceeds to step S37, and the keyword evaluation table 19 writes “+”, which is the third highest evaluation value in the four-level importance evaluation.

Further, when the calculated value P is smaller than 0.1, the ratio in which both the keywords A and B are "O" is the lowest, and in this case, the process proceeds to step S38 and the keyword evaluation table 19 has four important levels. The lowest evaluation value "-" is written.

In this evaluation method, the importance of two keywords is probabilistically evaluated based on the user's past judgment results, so that the keywords matching the information required by the user are estimated. You can

When evaluating the importance of a combination of a plurality of keywords, the importance of the keyword may be evaluated based on the temporal transition tendency of the user's past judgment results. The importance of a keyword may be evaluated by using the importance evaluation based on the above and the evaluation method (4) described above in combination.

Next, the processing contents executed by the service activation unit 20 based on the keyword evaluation table 19 created as described above will be described with reference to the flowchart of FIG.

The service activation unit 20 executes step S in FIG.
At 41, the keyword and its evaluation value are read from the keyword evaluation table 19. Then, it is determined whether or not the evaluation value of the keyword read in step S42 is "++" or more.

If the evaluation value of the keyword read from the keyword evaluation table 19 is lower than "++", the process directly returns to step S41 to read the next keyword, and if the evaluation value is "++" or more, the process proceeds to next step S43. The keyword is used as the information collection instruction table 22 (see FIG. 10).
(See) and returns to step S41.

When all the keywords in the keyword evaluation table 19 have been read in step S41, the process proceeds to step S44, in which the contents of the information collection instruction table 22 are sent to the information collection service execution section 14.

For example, the keyword evaluation table 19 is
When the service activation unit 20 has the configuration shown in FIG. 4, the evaluation value “+” is calculated from the keyword evaluation table 19.
When the keywords "+" or more are extracted, the information collection instruction table 22 displays the keyword "PB" as shown in FIG.
“X”, “TCIS”, and “○ × company + PBX” are written, and their contents are output to the information collection service execution unit 14.

The information collection service execution unit 14 executes the processing shown in FIG. 11 according to the instruction from the service activation unit 20 and collects the information including the keyword. First, Fig. 1
In step S51 of 1, the keyword registered in the information collection instruction table 22 is read, and in step S52, the information database 21 ₁ is searched using the keyword.

If the search result, that is, the information including the keyword registered in the information collection instruction table 22 exists in the information database 21 ₁ in step S53, the information is stored in the collection information file (not shown). .

When the search for the information about one keyword is completed, the process returns to step S51, the next keyword in the information collection instruction table 22 is read, and the information including the read keyword is searched from the information database 21 ₁ .

When the search of the information including the keyword from the information database 21 ₁ is completed, the process proceeds to S54 and the same search is performed for the other information databases 21 _{2 to} 21 _n .

When all the information including the estimated keywords is stored in the information collection file as described above, the information collection service execution unit 14 stores the information in the collection information file according to the access request from the user. The determined information is output to the user, the user's necessity / unnecessity of the information is determined, and the determination result is output to the user behavior monitoring unit 16. The user behavior monitoring unit 16 accumulates in the behavior database 17 the keywords included in the information presented to the user and the judgment result of the user regarding the information. As a result, the user's necessity / unnecessity judgment result for the information selectively collected based on the estimated keyword is also accumulated in the behavior database 17.

In the above embodiment, the interest for the presented information is
The result of the user's necessity / unnecessity is included in the information.
Save it in the behavior database in association with the keyword
User's past judgment result or past judgment
The importance of keywords is evaluated based on the tendency of migration.
Then, do not request access to the information collection service from the user.
When there is something, information including keywords with high evaluation values
Information database 21 ₁~ 21_nSearch from
I am trying to show you.

As a result, the user can obtain the necessary information without performing a keyword input operation. Furthermore, in this case, since information is searched for by all the keywords included in the information that the user has accessed in the past, information related to the keyword that the user does not directly specify is also presented at the same time, and the user can It becomes possible to collect desired information efficiently.

When the user uses the information collection service several times and the evaluation value of the keyword included in the information that the user has determined to be necessary is accumulated in the service control node 11, after that, the user is Since the network automatically searches and presents the information that is considered necessary for the user without inputting a keyword, the user needs to perform a troublesome key operation for searching the information from the database. Lost.

In the above embodiment, the case where the information collection service is used via the network has been described, but the present invention can be applied to services other than the information collection service.

Further, the information on the user's behavior stored by the user behavior monitoring unit 16 is not limited to the result of the judgment of necessity / non-necessity of information and the information on the keyword included in the information, and what kind of information the user has By accumulating information indicating whether a service was requested to the network in the past, when a user accesses the network, the service with the highest probability of being used by that user is automatically provided. Is also good.

In this case, the user can receive the desired service without performing the input operation for selecting the service each time. Further, the personal adaptive network control method of the present invention is not limited to the intelligent network system, but can be applied to a normal exchange system or LA system.
It can also be applied to networks such as N.

[0069]

According to the present invention, the network analyzes the behavior pattern of the user from the past behavior of the user with respect to the network and executes the processing based on the analysis result. Is reduced. For example, in the information collection service, information including keywords that are considered important to the user is presented, so the burden of input operation when searching for desired information from the database is reduced.

[Brief description of drawings]

FIG. 1A and FIG. 1B are explanatory diagrams of the principle of the present invention.

FIG. 2 is a system configuration diagram of an embodiment.

FIG. 3 is a configuration diagram of an action database.

FIG. 4 is a configuration diagram of a keyword evaluation table.

FIG. 5 is a flowchart illustrating a keyword evaluation method (1).

FIG. 6 is a flowchart illustrating a keyword evaluation method (2).

FIG. 7 is a flowchart illustrating a keyword evaluation method (3).

FIG. 8 is a flowchart illustrating a keyword evaluation method (4).

FIG. 9 is a flowchart showing processing of a service activation unit.

FIG. 10 is a diagram showing a configuration of an information collection instruction table.

FIG. 11 is a flowchart showing processing of an information collection service execution unit.

[Explanation of symbols]

 11 Service Control Node 14 Information Collection Service Execution Section 15 Information Collection Control Section 15 16 User Behavior Monitoring Section 17 Behavior Database 18 Analysis Section 19 Keyword Evaluation Table 20 Service Activation Section

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification number Internal reference number FI Technical location H04M 3/42 Z (72) Inventor Akira Hakuda 1015 Uedotachu, Nakahara-ku, Kawasaki City, Kanagawa Prefecture Fujitsu Stock In the company

Claims (5)

[Claims] 1. A behavior of a user of a network is monitored and stored, a behavior of the user is analyzed, a rule pattern is extracted, and a process corresponding to each user is executed based on the extracted rule pattern. A personal adaptive network control method characterized by the above. 2. The importance of the keyword included in the information is evaluated for each user based on the judgment result of necessity / unnecessity of each user for the presented information, and the keyword having a high importance is included. A personal adaptive network control method characterized by selectively collecting and presenting information to the user. 3. Information including a keyword having a high importance evaluated by evaluating the importance of the keyword included in the information based on the temporal transition of the judgment result of necessity / unnecessity of the user with respect to the presented information. A personal adaptive network control method, characterized by selectively collecting and presenting to the user. 4. A personal adaptive network control method, characterized in that the importance level of a keyword is evaluated by using the network control method according to claim 2 and the network control method according to claim 3. 5. The importance of the combination of a plurality of keywords contained in the information is determined based on at least one of the user's necessity / unnecessity judgment result and the time transition of the necessity / unnecessity judgment result for the presented information. A personal adaptive network control method, characterized in that information including a combination of a plurality of keywords evaluated and highly evaluated is selectively collected and presented to the user.