JPH08161174A - Information processing system and method thereof - Google Patents

Abstract

(57) [Summary] [Purpose] Understand the contents with minimum input. A knowledge base 22 having knowledge about a relationship between an input unit 1 for inputting information, a situation, a request according to the situation, and a response to the request in an information processing system
And an understanding unit 21 that infers a request corresponding to the information and the situation by searching the knowledge base 22 based on the input information, and an output unit 3 that outputs the request inferred by the understanding unit 21. And with.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system and method for processing information based on an understood situation.

[0002]

2. Description of the Related Art Conventionally, there is a system that interprets an input in natural language and infers the user's purpose or intention from this input. In order to realize this system, a natural language such as English is expressed by some computer (for example, comc
The system needs a mechanism that infers the goal behind a given input using a natural language parser (analysis part) for converting into ept dependency) and a knowledge base of goal plans.

There is also a system that does not allow input in natural language and performs a predetermined process by a command or a menu.

[0004]

In a conventional natural language processing system, a natural language input is translated into a computer expression and a goal is inferred from the computer expression based on a knowledge base of a goal plan expression. Was,
The adverse consequences of using knowledge of the goal plan could not be predicted.

Further, the cause-effect pairs in various situations do not necessarily have to represent the goal plan, but may represent other expressions such as a rule base.

Natural language input is too costly to construct a system, and natural language input that does not rely on voice input is not always easy and convenient, and is not desirable for the system. In some cases, an input method that does not depend on a natural language, such as a command method or a menu selection method, is preferable.

However, in the case of the input system which does not rely on natural language, the interpretation of the command is limited to the meaning fixedly defined in advance for the command, and the purpose or intention of the user is not taken into consideration. Further, in this system, when an erroneous content is input due to an operation error of the user or an erroneous calculation, the error cannot be corrected.

[0008]

In order to solve the above problems, the natural language processing system of the present invention provides an input means for inputting information, a situation, a request corresponding to the situation, and a response to the request. Based on the information input by the input means and a knowledge base having knowledge about the relationship between the knowledge base and an inference means for inferring a request corresponding to the information and the situation, and the inference means. Output means for outputting the inferred request.

According to another aspect of the present invention, the natural language processing system is provided with input means for inputting information, a situation, a request, a response to the request, a plan according to the request, and execution of processing. A knowledge base having knowledge of the relationship with the result, and a request inference means for inferring a request corresponding to the information by searching the knowledge base based on information from the input means, and corresponding to the request It comprises a formulating means for formulating a plan and a predicting means for predicting an execution result of the plan by simulation of the plan created by the formulating means.

According to another aspect of the present invention, with respect to the natural language processing system, the input means for inputting a command from the user, the history storage means for storing history information of the input by the input means, and the command A knowledge base having knowledge of, and an inference means for searching the knowledge base based on history information stored in the history storage means and inferring a command candidate to be input next from the input means,
Suggesting means for suggesting to the user the command candidates inferred by the inferring means.

According to another aspect of the present invention, the natural language processing system is provided with input means for inputting information from a user, knowledge about attributes of each user, and a request according to the attribute of the user. Based on the knowledge base having knowledge about the relationship with the response to the request, the information from the input means, and the attribute of the user who inputs the information, the knowledge base is searched to obtain the information and the user attribute. The inference means for inferring the corresponding request and the output means for outputting the request inferred by the inference means are provided.

According to another aspect of the present invention, in the natural language processing method, an input step of inputting information, a situation based on the information input from the input step, and a request corresponding to the situation. And an inference step of inferring a request corresponding to the information and the situation by searching a knowledge base having knowledge of the relationship with the response to the request, and an output step of outputting the request inferred by the inference step. Equipped with.

According to another aspect of the present invention, in the natural language processing method, an input step of inputting information, a situation, a request, and a response to the request based on the information from the input step, A request inference process for inferring a request corresponding to the information by searching a knowledge base having knowledge about the relationship between the plan according to the request and the execution result of the process, and formulating a plan corresponding to the request The method comprises a process and a prediction process for predicting an execution result of the plan by simulating the plan created by the process.

According to another aspect of the present invention, in the natural language processing method, an input step of inputting a command from a user, a history storing step of storing history information of the input by the input step, and the history. Based on the history information stored in the storage step, a knowledge base having knowledge about the command is searched, and an inference step of inferring a command candidate to be input next from the input step, and an inference step inferred by the inference step. A suggestion step of suggesting a command candidate to the user.

According to another aspect of the present invention, the natural language processing method includes an input step of inputting information from a user, information from the input step, and an attribute of the user who inputs the information. Based on the knowledge of the attributes of each user,
An inference step of inferring a request corresponding to the information and the user attribute by searching a knowledge base having knowledge about the relationship between the request corresponding to the user attribute and the response to the request, and the inference step inferred by the inference step. And an output process for outputting the request.

[0016]

In the present invention, the knowledge base is searched based on the input information, the request corresponding to the information and the situation is inferred, and the inferred request is output.

Further, according to another aspect of the present invention, the knowledge base is searched based on the input information, the request corresponding to the information is inferred, and the plan corresponding to the request is formulated and formulated. The execution result of the plan is predicted by simulation of the prepared plan.

According to another aspect of the present invention, the knowledge base is searched based on the stored history information, the candidate of the command to be input next is inferred, and the inferred command candidate is extracted. Suggest to the user.

According to another aspect of the present invention, the knowledge base is searched based on the attribute of the user who inputs the information, and the request corresponding to the information and the user attribute is inferred and inferred. Output request.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an information processing system 1 according to the present invention.
It is a block diagram which shows the hardware constitutions of an Example.

In FIG. 1, reference numeral 1 denotes an input unit for inputting information, for example, a keyboard for key-inputting characters and selection instructions, but not limited to this, a mouse for designating a position on the display. It may be a voice input device including a voice recognition device, an image input device including a character recognition device, a reception device that receives information from another device, or the like. You may use it for a specific purpose.
Furthermore, information generated by other processing in this device may be input here.

Reference numeral 2 denotes a CPU, which performs arithmetic operations for various processes, logical judgments, etc., and controls each constituent element connected to the bus 6.

An output unit 3 is, for example, a display device such as a CRT or a liquid crystal display, but in addition to this, an audio output device for synthesizing output information by voice and outputting it, or a printer for printing out. , A transmitting device that transmits information to another device, or two or more of these may be provided and selectively used. Further, the output here may be used as an input to other processing in the apparatus.

Reference numeral 4 denotes a program memory, which is a memory for storing a program for control by the CPU 2 including a processing procedure described later in the flowchart. The program memory 4 may be a ROM or a RAM loaded with a program from an external storage device or the like.

Reference numeral 5 denotes a data memory, which stores data generated by various processes and also stores knowledge of a knowledge base described later. The data memory 5 is, for example, a RAM, but the knowledge of the knowledge base is stored in a non-volatile external storage medium.
It should be loaded prior to processing, or referenced whenever necessary.

Reference numeral 6 denotes a bus for transferring an address signal indicating a constituent element to be controlled by the CPU 1, a control signal for controlling each constituent element, and a data exchanged between the constituent devices. is there.

FIG. 2 is a functional block diagram showing the basic configuration of the information processing apparatus according to the embodiment of the present invention.

In FIG. 1, an input unit 1 inputs a command from a user, a device operation state, surrounding environment information and the like. Reference numeral 21 is an understanding unit that refers to the knowledge base 22 based on the information input from the input unit 1 to understand the input information, the current situation, and the like, and generates necessary output information. Reference numeral 22 is a knowledge base having knowledge of causes and effects. The information generated by the understanding unit 21 is output to the output unit 3 and provided to the user, a specific device, or the like.

Next, as an example in which the operation of the device is treated as a command, an example of controlling the driving of the vehicle will be described.

In this example, the sensor detects and inputs the operating conditions of the user's accelerator, brake, steering wheel, etc., the presence / absence of vehicles in front of and behind, the traveling speed of the vehicles, and the inter-vehicle distance. To do. Also, as output,
Control information for executing deceleration, acceleration, lighting of lights, etc. is output.

FIG. 3 is a block diagram of the understanding unit 21 in this embodiment. Reference numeral 31 is a goal detection unit that detects a user's intention, 32 is an inference unit that infers an action necessary to achieve the detected goal, and 33 is an operation performed by the inference unit 32. A simulation unit 34 for checking whether or not there is an adverse effect is an execution unit. FIG. 4 is an example of a rule base in this embodiment.

First, when the user depresses the brake, it is inferred that deceleration is necessary according to the rule (a) of FIG. Further, when it is detected that the vehicle in front of the vehicle is braked as the surrounding situation, it is inferred from the rule (d) that the inter-vehicle distance from the vehicle in front is shortened, and the rule (c) is further added. It is inferred that the vehicle-to-vehicle distance from the preceding vehicle must be increased, and that the rule (b) requires that the vehicle speed be reduced, that is, decelerated, for that reason. This is consistent with the inference by (a). From the above reasoning, it is understood that the reason why the brake pedal is applied is that the preceding vehicle has applied the brake. Also, the goal of slowing down is obtained. Here, since deceleration is required, according to rule (e),
If you have a car behind you, turn on the hazard lights,
Slow down as slowly as possible. In other cases, the brake will be decelerated by the required value. Further, according to the rule (f), when the preceding vehicle starts accelerating during deceleration, the inter-vehicle distance with the preceding vehicle increases, and if the inter-vehicle distance increases, stop deceleration and start accelerating. You can

Next, a system for creating a letter by selecting a required item by menu selection will be described.

FIG. 5 is a diagram showing an example of a menu screen for inputting the information of the sender of a letter. The name of the sender,
Enter the affiliation, etc.

FIG. 6 is a diagram showing an example of a menu screen for inputting the information of the person to whom the letter is addressed. The name, affiliation, etc. of the person to whom the letter is addressed are input.

In these inputs, how to write a name is
The order of writing surnames differs depending on the country. Therefore, the system is provided with a model for correcting an input error by the user. For example, in FIG. 6, John, which should be the first name, has been input as the family name. Therefore, this is corrected. In addition, the system tries to supplement the unentered information regarding John by referring to a database or the like.

FIG. 7 is a diagram showing an example of a part of SDB which is a database relating to individuals.

Since there are two Johns in the SDB of FIG. 7, the system performs the confirmation process shown in the flowchart of FIG.

First, in step S81, the user is asked which John he is.

When the user selects John Smith in step S82, John Smith is supposed to belong to the University of Rochester in the SDB, but here Duke is input together with John, so in step S83. ,
The system asks the user if John has moved to Duke University.

If the user's answer is yes, the process proceeds to step S84, where John Smith's affiliation in SDB is changed to Duke University. If the user's answer is the user's mistake, the concept is corrected in step S85. If the user's answer is that the user's answer is a temporary movement to a certain time, the fact is entered in the SDB in step S86.

In the above, if there is only one John who belongs to Duke University in the SDB, the processing of FIG. 8 is omitted and the person is confirmed or the information of the person is input. You may make it a user confirm.

If no communication / relationship has been found between the sender and the person at the destination, the system judges that it is the first letter and displays the menu of FIG.

FIG. 9 is a menu screen for selecting the purpose of the letter.

In this menu, the user has selected 1, ie, the user wants to visit John. Then, the system displays the menu of FIG. 10 to get the visit subgoal from the user. even here,
Suppose you choose 1.

FIG. 11 is a menu for entering information about visit details. The shaded part in the display is the part inferred by the system. Regarding the date, the visit is also inferred to be 1993, assuming the current time is April 10, 1993. Furthermore, if the user inputs 7 days as shown in FIG. 11, the closest date is 5
It may be inferred that it is the month, and then the user need only enter 7.

It is inferred that the visit destination is likely to be John's place, and John's institution, Duke Uni
versity is displayed. Of course, the user can rewrite the information input on the system side.

As shown in FIG. 11, when the user answers "y" in response to a request from the system, the menu shown in FIG. 12 is displayed. Here, it is assumed that the user wants John to reserve an accommodation and selects 1.

Based on the above input, a letter as shown in FIG. 13 is output.

In the above example, the system is not limited to the general goal of looking for information in inferring why the user wants to visit the destination person, but also why the user can get information from the other party. I also infer from the point that I believe that. The system then searches for a relationship between the user's work knowledge and the other's work knowledge.

Further, the system realizes a function of inferring and planning using the contents of the previous communication record (existing in the operation history of the user).

For example, when making an appointment for the first letter, the contents of the previous letter are checked, the overlapping schedules are checked, and the visit is made at a specific time (a time that coincides with the time of the meeting, etc.). Infer the reason for. In the case of a reply to a letter from the other party, the system establishes a relationship with the other party using the history, fetches the latest letter from the database, and the content of the letter gives the user's reply (for example, is the date and time accepted? Or not).

Next, an example in which the system prompts the user for the remaining commands when the commands are input halfway will be described.

If there is one possible command, for example "Sav" was entered by the user, then the only corresponding command is specified as "Save".

When there are a plurality of possible commands, for example, when the user inputs "co",
The complete commands are "compile", "copy", "continue", etc.

When there are a plurality of possible commands, when the system keeps a history of user command inputs, for example, when the user has already saved the file, the above "co" is It is inferred that it will be "compile" after the save.

Next, an example of creating a letter based on natural language input will be described. The details of this example are described in Japanese Patent Application No. 6-138887 filed by the applicant of the present application.

As shown in FIG. 14, "Reply to John's
letter accepting to meet on 10thMay at 3 pm to dis
"cuss on PDA systems." is input, the concept analysis unit (C that analyzes the natural language input and outputs the corresponding concept)
A) outputs the concept of C1 in the figure to the understanding unit 21 as the analysis result of this input.

Here, the processing by the understanding main processing unit will be described assuming that there is no processing target of the pre-understanding processing unit of the understanding unit 21.

First, first, the actor included in the information is searched for. As shown in a) of FIG. 15, here, the sender is Aruna Rohra and the receiver is John Smith.
Also, the type of letter is reply. In addition, it is assumed that the database has no content related to it so far. However, the contents can be obtained from the input. Also, the sender is expecting a reply and can guess the letter of the recipient from C3 obtained from the input.

Then, when the understanding main processing unit processes C3, the goal is inferred as shown in d) of FIG. 15, and the chain structure inference as shown in FIG. 16 is performed. Next, make a plan for understanding. That is, a task network having the structure shown in FIG. 17 is constructed. Furthermore, the post-understanding processing unit supplements the shortage of users in the plan of FIG. 17 to obtain FIG. Then rebuild the letter from the above plan for confirmation.

Next, the sender's reply is created from the input (C2 obtained from the input). As shown in FIG. 15 e) for the previous letter CD obtained as described above,
S to discuss PDA system S to meet and accept MAINGOAL Accept MAINPLAN, S For PLAN (for example, after May 8th), 3:00 pm on May 10th, which is a more specific date and time, is designated. Based on the above, the letter of FIG. 19 is created.

Further examples of other letters will be described. In this example, the previous contents can be obtained from the database. Hereinafter, differences from the above embodiment will be mainly described.

As shown in FIG. 20, "Reply to John's
letter confirming the meeting time, 19th May at 3p
m. Also, request him to arrange for pickup. "is input, the concepts of C1 and C4 are output from the CA 31.

Understand that the main processing unit obtains the related contents from the database, visits after May 10, and responds to John's letter. Also, as shown in FIG. 21 d), the previous letter is retrieved from the database. As shown in FIG. 21e), S S to meet and accept MAINGOAL Accept MAINPLAN. These things are "... co
It is inferred from the input "nfirming the meeting time, 19th May at 3pm." S For the PLAN, a more specific date and time, 3:00 pm on May 19, is designated. Also, as the sender's letter, in addition to the reply created as described above, another request is added (a request to be carried is obtained from "request him to arrange for pickup"). Thereafter, the same processing as in the above-described embodiment is performed, and the letter of FIG. 22 is created.

Further examples of other letters will be described. In this example, the system cannot infer the recipient's letter from the current input, and requests the user for input. Hereinafter, the differences from the above-described embodiments will be mainly described.

As shown in FIG. 23, "Reply to John's
When "letter." is input, the concept of C1 is output from CA31.

The understanding main processing unit obtains the related contents from the database, finds that the visit will be made after May 10, and confirms that the user should reply to John's letter. Also, as shown in FIG. 23 d), the previous letter is retrieved from the database. In e) of FIG. 23, the system is unable to guess the recipient's letter, as the current entry does not specify anything, and prompts the user to enter the contents of this letter, for example by OCR. . Then, the content of the letter as shown in e) of FIG. 23 is input. Thereafter, the processing of FIG. 24 is executed in the same manner as in the above-described embodiment, and the letter of FIG. 25 is created.

Next, an example in which the operation of the device is requested by natural language will be described.

FIG. 26 is an example of knowledge of the knowledge base regarding commands in this embodiment, and each process and the command input operation procedure for the process are stored in the goal plan format.

FIG. 27 is a diagram showing a concrete input example and a plan to be executed. As shown in FIG.
e size and font of the letter to bold and 12 poin
When "t." is input, a procedure for making the style bold and the size 12 points is searched based on the knowledge of FIG. From the format of the input sentence, the user requests the device to execute the operation based on this command, so as shown in FIG. 27, the procedure obtained by the search is used as the execution plan and this plan is executed. To do.

As long as the functions of the present invention are executed, the processing is performed by supplying a program to the device or the system, whether it is a single device or a system composed of a plurality of devices. Needless to say, the present invention is applied even in such cases.

[0074]

As described above, according to the present invention,
Optimal response can be achieved from the minimum input given.

Further, even if there is an error in the input information, the error can be corrected and a response can be made.

Further, it is possible to avoid the risk of executing an inconvenient process such as an erroneous command.

[Brief description of drawings]

FIG. 1 is a block diagram showing a hardware configuration of an information processing system according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a basic configuration of an information processing system.

FIG. 3 is a diagram showing a block configuration of an understanding unit.

FIG. 4 is a diagram showing an example of a rule base.

FIG. 5 is a diagram showing an example of a menu for inputting a sender name.

FIG. 6 is a diagram showing an example of a menu for inputting a recipient's name.

FIG. 7 is a diagram showing an example of a personal database.

FIG. 8 is a flowchart of confirmation processing.

FIG. 9 is a diagram showing an example of a menu for inputting the purpose of a letter.

FIG. 10 is a diagram showing an example of a menu for inputting a subgoal.

FIG. 11 is a diagram showing an example of a menu for inputting detailed information on a visit.

FIG. 12 is a diagram showing an example of a menu for inputting an additional request.

FIG. 13 is a diagram showing an example of an output letter.

FIG. 14 is a diagram illustrating an example of CA processing in creating a letter.

FIG. 15 is a diagram showing a process of creating a letter.

FIG. 16 is a diagram showing an example of inference of a chain structure.

FIG. 17 is a diagram showing an example of a task network.

FIG. 18 is a diagram showing an example of a task network that compensates for a shortage.

FIG. 19 is a diagram showing an output example of a letter.

FIG. 20 is a diagram illustrating an example of CA processing in creating a letter.

FIG. 21 is a diagram showing a process of creating a letter.

FIG. 22 is a diagram showing an output example of a letter.

FIG. 23 is a diagram illustrating an example of CA processing in creating a letter.

FIG. 24 is a diagram showing a process of reconstructing a letter.

FIG. 25 is a diagram showing an output example of a letter.

FIG. 26 is a diagram showing an example of a knowledge base regarding commands.

FIG. 27 is a diagram showing an input side and a plan to be executed.

[Explanation of reference symbols] 1 input unit 2 CPU 3 output unit 4 program memory 5 data memory 6 bus 21 sentence creation unit 22 knowledge base 23 output generation unit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location 9194-5L G06F 15/403 330 C

Claims (50)

[Claims] 1. A means for inputting information, a situation, a knowledge base having knowledge about a relationship between a request corresponding to the situation and a response to the request, and information based on the information input by the input means. And an inference means for retrieving the knowledge base to infer a request corresponding to the information and the situation, and an output means for outputting the request inferred by the inference means. . 2. The information processing system according to claim 1, wherein the inference means further infers a request whose response is a request once obtained by the inference. 3. The information processing system according to claim 1, wherein the input unit includes a selection unit that selects one option from a plurality of options. 4. The information processing system according to claim 1, wherein the input unit includes an operation unit for operating a predetermined mechanism. 5. The information processing system according to claim 1, further comprising a processing unit that executes a process according to a request output from the output unit. 6. The information according to claim 1, further comprising a situation obtaining means for obtaining information about a situation, wherein the inference means makes an inference based on the situation obtained by the situation obtaining means. Processing system. 7. The information processing system according to claim 6, wherein the situation acquisition means includes an observation means for observing and inputting environmental information around the information processing system. 8. The condition obtaining means includes: history storage means for storing a history of input by the input means; and reference means for referring to stored contents of the history storage means. Information processing system described. 9. An input unit for inputting information, a situation, a request, a response to the request, a knowledge base having knowledge about a relation between a plan according to the request, and an execution result of processing, and the input unit. Based on the information from the knowledge base, the request reasoning means for inferring the request corresponding to the information, the formulating means for formulating a plan corresponding to the request, and the plan prepared by the formulating means An information processing system, comprising: a prediction unit that predicts the execution result of the plan by the simulation. 10. The information processing according to claim 9, further comprising: a determining unit that determines a process to be performed in order to avoid the unfavorable result when the unpredictable result is predicted by the predicting unit. system. 11. A notifying means for notifying the user of the predicted execution result.
The information processing system described in. 12. The apparatus according to claim 10, further comprising execution means for executing the processing determined by the determination means.
The information processing system described in. 13. The input means has one of a plurality of options.
The information processing system according to claim 9, further comprising a selection unit that selects one option. 14. The information processing system according to claim 9, wherein the input unit includes an operation unit for operating a predetermined mechanism. 15. The information according to claim 9, further comprising a situation obtaining means for obtaining information about a situation, wherein the inference means makes an inference based on the situation obtained by the situation obtaining means. Processing system. 16. The information processing system according to claim 15, wherein the situation acquisition means includes an observation means for observing and inputting environmental information around the information processing system. 17. The status acquisition means includes history storage means for storing a history of input by the input means, and reference means for referring to stored contents of the history storage means. Information processing system described. 18. An input unit for inputting a command from a user, a history storage unit for storing history information of the input by the input unit, a knowledge base having knowledge about commands, and a history storage unit stored in the history storage unit. An inference means for retrieving the knowledge base based on history information and inferring a command candidate to be input next from the input means; and a suggesting means for suggesting a command candidate inferred by the inference means to a user. An information processing system characterized by comprising. 19. The input means is means for inputting a command as a character string, the history storage means stores a character string being input by the input means, and the inference means is
The information processing system according to claim 18, wherein a possible character string command is inferred from the character string being input. 20. The history storage means stores at least one previously input command, and the knowledge base is
The information processing system according to claim 18, wherein one command and a command that can be input subsequent to the command are stored. 21. The input means is means for inputting a command as a character string, the history storage means further stores a character string being input by the input means, and the inference means is the history storage. A command stored in the knowledge base as a command that can be input subsequent to at least one previously input command stored in the means, and a possible character string command from the input character string The information processing system according to claim 20, wherein: 22. An input means for inputting information from a user, a knowledge base having knowledge about attributes of each user, a request corresponding to the attribute of the user, and knowledge about a relationship to the request, An inference unit that infers a request corresponding to the information and the user attribute by searching the knowledge base based on the information from the input unit and the attribute of the user who has input the information, and inferred by the inference unit. And an output unit for outputting the request. 23. The information processing system according to claim 22, wherein the knowledge base stores a history of information input by a user as an attribute of the user. 24. The information processing system according to claim 23, wherein the inference means refers to the schedule information of the user as a history of input information from the user. 25. The information processing system according to claim 23, wherein the inference unit refers to user communication information as a history of input information from the user. 26. A knowledge base having knowledge about an input step of inputting information, a situation, a request corresponding to the situation, and a response to the request, based on the information input by the input step. And an output step of outputting a request inferred by the inference step, and an inference step of inferring a request corresponding to the information and the situation. 27. The information processing method according to claim 26, wherein the inference step further infers a request whose response is a request once obtained by the inference. 28. The input step comprises one of a plurality of options.
27. The information processing method according to claim 26, further comprising a selection step of selecting one option. 29. The inputting step includes an operating step for operating a predetermined mechanism.
Information processing step described in. 30. The information processing method according to claim 26, further comprising a processing step of executing processing according to the request output from the output step. 31. The information according to claim 26, further comprising a situation obtaining step of obtaining information about a situation, wherein the inference step makes an inference based on the situation obtained by the situation obtaining step. Processing method. 32. The information processing method according to claim 31, wherein the situation acquisition step includes an observation step of observing and inputting environment information around a system that executes the information processing method. 33. The method according to claim 31, wherein the situation acquisition step includes a history storage step of storing a history of input by the input step, and a reference step of referring to stored contents of the history storage step. Information processing method described. 34. An input process for inputting information, and a relationship between a situation, a request, a response to the request, a plan according to the request, and an execution result of processing based on the information from the input process. A request reasoning step of searching a knowledge base having knowledge and inferring a request corresponding to the information, a formulating step of formulating a plan corresponding to the request, and a plan of the plan prepared by the formulating step. An information processing method comprising: a prediction step of predicting an execution result of. 35. The information processing according to claim 34, further comprising a determining step of determining a process to be performed in order to avoid the unfavorable result when the unpredictable result is predicted by the predicting step. Method. 36. A notification step of notifying the user of the predicted execution result.
4. The information processing method described in 4. 37. The method according to claim 35, further comprising an execution step of executing the processing determined by the determination step.
Information processing method described in. 38. The input step comprises one of a plurality of options.
The information processing method according to claim 34, further comprising a selection step of selecting one option. 39. The input step includes an operation step for operating a predetermined mechanism.
Information processing method described in. 40. The information according to claim 34, further comprising a situation obtaining step of obtaining information about a situation, wherein the inference step makes an inference based on the situation obtained by the situation obtaining step. Processing method. 41. The information processing method according to claim 40, wherein the situation acquisition step includes an observation step of observing and inputting environmental information around the information processing method. 42. The method according to claim 40, wherein the situation acquisition step includes a history storage step of storing a history of input by the input step, and a reference step of referring to stored contents of the history storage step. Information processing method described. 43. An input step of inputting a command from a user, a history storage step of storing history information of the input by the input step, and knowledge of the command based on history information stored in the history storage step. And a suggestion step of suggesting a command candidate inferred by the inference step to the user by inferring a command candidate to be input next from the input step. A characteristic information processing method. 44. The input step is a step of inputting a command as a character string, the history storing step stores a character string being input by the input step, and the inference step is
The information processing method according to claim 43, wherein a possible character string command is inferred from the character string being input. 45. The history storing step stores at least one previously input command, and the knowledge base is
The information processing method according to claim 43, wherein one command and a command that can be input subsequent to the command are stored. 46. The input step is a step of inputting a command as a character string, the history storage step further stores a character string being input by the input step, and the inference step is the history storage. A command stored in the knowledge base as a command that can be input subsequent to at least one previously input command stored in the process, and a possible character string command from the character string being input The information processing method according to claim 45, wherein: 47. Based on the input step of inputting information from the user, the information from the input step, and the attributes of the user who input the information, the knowledge about the attribute of each user and the attribute of the user An inference step of inferring a request corresponding to the information and the user attribute by searching a knowledge base having knowledge about the relationship between the request and the response to the request, and outputting the request inferred by the inference step An information processing method comprising an output step. 48. The information processing method according to claim 47, wherein the knowledge base stores a history of information input by a user as an attribute of the user. 49. The information processing method according to claim 48, wherein the inference step refers to user's schedule information as a history of input information from the user. 50. The information processing method according to claim 48, wherein the inference step refers to user communication information as a history of input information from the user.