JPH0247965A - Communication service semantic understanding system - Google Patents

Abstract

PURPOSE:To prevent improper operation by decoding the meaning of it when the user instructs a communication service and evaluating whether or not the service is proper and outputting the result. CONSTITUTION:A user state extraction means 2 is started at the production of communication service request and the user state is extracted based on a user state extraction rule 3 from a data from a data group storage section 1 and the result is outputted to a service evaluation means 4. On the other hand, when the content of the communication service requesting the user is inputted, it is set to a communication service type input section 8 and the data is made correspondent according to an information flow.service corresponding rule 7 in an information flow.service corresponding means 6 to generate an information flow data and the result is fed to the service evaluation means 4. Based on an information flow model 5, the means 4 evaluates whether the combination of the user status information and the service flow information given from the means 2, 6 is proper. Then when the combination is discriminated to be improper, an inquiry means 10 inquires about it to the user.

Description

[Detailed Description of the Invention] [Summary] This invention relates to a communication service meaning understanding method that determines whether various communication services are appropriate depending on the situation of a user using a communication system. The purpose is to provide a semantic understanding method for communication services that notifies users when inappropriate instructions are given, based on user state extraction rules from a data group storage that stores data related to the user's situation and environment. and an information flow/service response means that outputs a communication pattern representing the information flow based on the information flow/service response rule from the type input section of the communication service requested by the user. and a service evaluation means for evaluating whether or not the communication pattern from the information flow/service handling means is compatible with each state of the user extracted by the user state extraction means, based on an information flow model, The system is configured to generate an output to inquire the user when the service evaluation means determines that the service is non-conforming.

[Industrial Application Field] The present invention relates to a communication service meaning understanding method for determining whether various communication services are appropriate depending on the situation of a user using a communication system.

In recent years, the technology of both terminal devices and switching devices constituting communication systems has made remarkable progress, and many functions have become executable, and services for users are diversifying and increasing in number.

In order to perform such a large number of service functions, a user has to ask the system to accept the service request, but conventionally, the communication system simply accepts the request as is, and the service depends on the user's situation. Even if it is inappropriate in light of the above, it is accepted, and inconveniences may occur when the service is executed. Therefore, it is desired to be able to understand and evaluate the content of communication service requests.

[Prior Art] Conventionally, the meaning of the operation of a communication system is not determined and the evaluation of whether the operation is appropriate or inappropriate for the user is not performed, and everything is left up to the user. Therefore, the communication system performs the service according to the user's instructions without making any semantic evaluation of its own operation.

[Problems to be Solved by the Invention] As mentioned above, conventional systems do not perform any evaluation, so even if a user instructs a service request, the operation desired by the user may not be performed, or an operation not desired by the user may occur. A problem arises in that there are cases where the appropriate action is taken.

That is, if the user forgets an instruction, gives an instruction that is contrary to a pre-registered instruction, or accidentally gives a semantically inappropriate instruction, the operation that the user wants to perform cannot be executed.

Problems like this would not have been such a problem if the instructions given by the user were simple, as in the past, but as services become more sophisticated and diversified in the future, the efficient use of communication systems and the This is an increasingly important issue as the need to support users of various services increases.

SUMMARY OF THE INVENTION An object of the present invention is to provide a semantic understanding method for communication services that understands the meaning of an instruction from a user and notifies the user when a semantically inappropriate instruction is given or when there is no instruction.

[Means to solve the problem] FIG. 1 is a diagram showing the basic configuration of the present invention.

In FIG. 1, 1 is a data group storage unit that stores data related to users and the system, 2 is a user state extraction means, 3 is a user state extraction rule, 4 is a service evaluation means, 5 is an information flow model, and 6 is information Flow and service response methods, 7
Reference numeral 8 indicates an information flow/service correspondence rule, 8 indicates a communication service type input section representing the content of the communication service requested by the user, 9 indicates a registration means, and 10 indicates an arrangement means.

The present invention includes knowledge data for understanding the meaning of services such as user state extraction rules, information flow/service correspondence rules, and information flow models, and is capable of interpreting the meaning when a user instructs a communication service. It evaluates and outputs whether or not the service is appropriate in light of the situation.

[Function] The data group storage unit 1 shown in FIG. 1 stores a pre-registered data group indicating the status of the user's schedule, communication history, etc. When activated, the user status is extracted from the data representing the situation from the data group storage unit 1 based on the user status extraction rule 3 and outputted to the service evaluation unit 4.

On the other hand, when the user inputs the content of the requested communication service, it is set in the communication service type input section 8, and the data is associated in the information flow/service correspondence means 6 according to the information flow/service correspondence rules 7. Information flow data representing the type of information flow (communication pattern) of the requested service is generated and supplied to the service evaluation means 4. In the service evaluation means 4, based on the information flow model 5, the user state information and service flow information input from the user state extraction means 2 and the information flow/service handling means 6 are determined to be appropriate (permissible) Ml.
Evaluate whether it is a combination or an inappropriate combination.

Through this evaluation, even if a semantically inappropriate service is instructed due to a user's error, it can be detected, and by confirming the instruction to the user by the intermediary means 10, execution of an inappropriate service can be prevented. It can be avoided. If it is evaluated as appropriate, the service is registered in the registration means 9 in order to execute the service.

With this kind of evaluation, even when there is no instruction from the user,
The system can autonomously select appropriate services. That is, this can be achieved by selecting a communication flow that matches the user status using the information flow model 5 and selecting a corresponding service from the information flow/service correspondence rules 7.

[Example] FIG. 2 shows the configuration of an embodiment of the present invention.

In the figure, 20 is a terminal, 21 is a network (indicated by NW) that constitutes the communication path of the exchange, 22 is a call processing unit for controlling the connection of the exchange, 23 is an additional processor that processes various services for users, Reference numerals 24 and 25 represent an interface (indicated by IF), 26 a magnetic disk in which the semantic understanding knowledge data 261 is stored, 27 a CPU (central processing unit), and 28 a memory.

This embodiment is an example in which the communication service semantic understanding method is implemented in the incoming call control of an exchange, and it goes without saying that the communication service semantic understanding method can be implemented for the terminal 20.

Semantic understanding knowledge data 26 stored in the magnetic disk 26
1 is the user state extraction rule 3 shown in FIG. It consists of an information flow model 5 and information flow/service correspondence rules 7.

In the incoming call control, the user registers in advance the operation to be performed when a call arrives as a condition-service pair, and the operation is performed according to the condition-service pair when the call arrives.

When the user makes a call from the terminal 20 and makes a specific dial, the network 21 is switched and connected to the additional processor 23 under the control of the call processing section 22.Next, the user makes a call processing instruction as shown in FIG. When a call processing instruction with the content shown is input, it is stored in the call processing instruction requesting section 281 in the second memory. Similarly, the user can
When a schedule table as shown in FIG. 4 is input, its contents are stored in the schedule table 282 of the memory 28.
is stored in

The magnetic disk 26 stores user status extraction rules as shown in FIG. 5, information flow/service correspondence rules as shown in FIG. 6, and information flow model as shown in FIG.

In this embodiment, from these knowledge data and the schedule table, it is checked whether or not the manually entered call processing instructions are semantically appropriate, and if inappropriateness is detected, confirmation is made to the user. These operations are carried out by the service. Semantic analysis program 283
FIG. 8 shows a flowchart of the service semantic analysis process.

Next, a specific example of the operation will be described with reference to FIGS. 2 and 8. Now, it is assumed that the system has the knowledge shown in Figures 5, 6, and 7, and the schedule table shown in Figure 4 is registered, and the user connects to the exchange for a service request using the steps described above. Assume that the call processing instruction shown in FIG. 3 is input by performing the following steps.

The system operates according to the processing flow shown in Fig. 8. First, the first instruction is from 9:00 to 11=00.
When this user (the user instructing call processing) receives an incoming call from a caller named "Nakaatsu," the user would like to receive a "message reception service." Note that the caller's identification is performed, for example, on the 15DN digital network, and can be known.

In response to this instruction, all event names corresponding to the time specified by the call processing instruction are read from the schedule table 282 (FIG. 8 81). In this case, as shown in FIG. 4, time 9 regarding other users: When the event name corresponding to 00 to 11:00 is read out, "meeting" is obtained.

Next, the read event name is compared with the conditions of the user state extraction rule, and the user state for each event name is determined (8
2) In this example, when the user state corresponding to the condition "meeting" is extracted according to the user state extraction rule (FIG. 5) in the semantic understanding knowledge data 261, the state "busy" is obtained.

On the other hand, the information flow pattern corresponding to the service salt specified by the call processing instruction stored in the call processing instruction request unit 281 is determined from the information flow and service correspondence rules (
83), in this example, for the service salt ``Receive message'', ``Standby, one direction'' is obtained as the information flow pattern from the information flow/service correspondence rules (Fig. 6).

Next, the combination of each user state and information flow pattern obtained through the above processing is compared with the information flow model,
In this example, the semantic evaluation of the combination of "busy" and "waiting, one way" determined above is determined from the information flow model in Figure 7. When asked, "O" (indicating that it is appropriate) is obtained, so this call processing instruction is registered in the call processing instruction registration unit 284 (88) without asking the user, and the process ends. .

The processing of steps 81 to 84 in FIG. 8 is also performed for the second instruction. This instruction is from 12:00 to 17:00.
When this user receives an incoming call from a caller named "Kawasaki" at 00, the user requests that the call be forwarded.

The transfer destination is the location shown in the schedule table (
For example, a conference room).

When the event name corresponding to the time 12:00 to 17:00 corresponding to this instruction is read from the schedule table, there is no event in the schedule table for the period 12:00 to 14:00, and 14:00 to 17:00 :00:
You can get "visitors".

Next, according to the user state extraction rule (FIG. 5), the user state corresponding to the condition "visitor" is "busy", which is the same as "meeting". For no event, the user status is ``normal''.For the zero-order service salt ``transfer'', from the information flow/service correspondence rules (Figure 6), the information flow pattern is ``immediate,'' You can get ``both directions''.

As a result of the above, the requested "busy" and "normal" and "
When the semantic evaluation of the combination of "immediate, bidirectional" is obtained from the information flow model (Figure 7), "O" is obtained for the combination of "normal" and "immediate, bidirectional";
For busy J, the mark is "×" and it is detected that it is semantically inappropriate. Therefore, time 14:00-17+
o O presents the user with a schedule that there will be a visitor and asks for confirmation of this call processing instruction (8687
), if the user approves, this call processing instruction is registered in the call processing instruction registration section 284 (88), and if the user cancels, the process ends without registration.

As shown in the above example, in this embodiment, a semantically inappropriate communication service instruction can be detected, and by taking appropriate processing at that time, inappropriate operations can be avoided.

[Effects of the Invention] According to the present invention, the communication system can select and execute an appropriate service after evaluating the meaning of each service. Further, when it is detected that the instructed service is inappropriate, it is possible to take measures such as confirming the user's intention, thereby reducing semantically incorrect operations.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the basic configuration of the present invention, FIG. 2 is a diagram showing the configuration of an embodiment of the invention, FIG. 3 is a diagram showing an example of call processing instructions, FIG. 4 is a diagram showing an example of a schedule table, and FIG. Figure 5 shows an example of user state extraction rules, Figure 6 shows information flow/service correspondence rules, Figure 7 shows an example of an information flow model, and Figure 8 shows service semantic analysis processing. FIG. In Figure 1, 1: Data group storage unit 2: User state extraction means 3: User state extraction rules 4: Service evaluation means 5: Information flow model 6: Information flow/service handling means 7: Information flow/service Correspondence rule 8: Communication service type input section 9: Registration means 10: Arrangement means Figure 7 shows

Claims (1)

[Claims] In a communication service semantic understanding method that accepts a communication service request, determines its suitability, and outputs the request, a user state extraction rule is provided from a data group storage unit (1) that stores data related to the user's situation and environment. A user status extraction means (2) that extracts a user status based on (3), and a communication service type input unit (8) requested by the user, which extracts the information flow based on the information flow/service correspondence rule (7). an information flow/service response means (6) that outputs a communication pattern representing the information flow/service response means (6); and the communication pattern from the information flow/service response means (6) in each state of the user extracted by the user state extraction means (2). The information flow model (
A communication service meaning understanding method, comprising: a service evaluation means (4) that evaluates based on 5), and generates an output to inquire to a user when the service evaluation means (4) determines that the service is non-conforming.