JPH08292864A - User interface and customizing method for application program - Google Patents

Abstract

PURPOSE: To automatically set default values of respective application programs according to a user by setting the object at an output destination to a found prescribed value by inference according to an inference rule by using an input instance that a user inputs and a correspondence relation table regarding the same user. CONSTITUTION: An inference device 5 consists of a control part 8, a learning part 9, an inference part 10, and a storage part 11, in which correspondence relation tables of input destinations and output destinations as to objects previously specified by users among objects of respective displayed application programs are registered, user by user. The prescribed values of the object at the output destination is inferred according to the inference rule by using the input instance that the user inputs and its correspondence relation table and the object at the output destination is set to the prescribed value found by the inference. Then when the user changes the prescribed value of the object at the output destination to a prescribed value different from the inference result, the inference rule is learnt.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for customizing a user interface and an application program.

[0002]

2. Description of the Related Art As is well known, a graphical user interface (GUI) that interactively executes an operation for each application program via a display screen is widely used as an interface of various systems.

By the way, when such a GUI is used by a plurality of users, the commands and input values used for each user often differ greatly due to the difference in the usage environment of each user. For example, there are some people who usually use only one set of printouts, and some who usually use two sets of printouts, even if they use the same document creation software, depending on the work content of the user. In such a case, in the former case, it is preferable that the setting value of the number of printed sheets is 1 without specifying anything, and in the latter case, the setting value of the number of printing sheets is preferably 2 without specifying anything.

These initial values are default values (default values)
is called. It is desirable that the default value is not limited to the dialog in which numbers are entered as described above, but the value most used by each user is automatically given to the menu or the exclusive switch.

As a method for realizing this, conventionally, a method of holding a value last used by the user, a method of using a value preset by the user, and a method of determining the most frequently used frequency based on data recorded the frequency of use A method of setting a high value is known.

In these conventional methods, when the value of one object (numerical value input field, menu, switch) influences the default value of another object, the user selects some customization language. It is necessary to create the application program after describing the dependency relationship by using it or predicting that the dependency relationship will occur in advance at the program creation stage.

However, describing the default behavior in an application program in a language for customization is extremely troublesome for a general user. On the other hand, even when learning is performed using the dependency relationships built into the system, the usage patterns of users vary widely,
It is extremely difficult to predict all possible dependencies and incorporate them into the system in advance as a learning target.
Also, when many dependencies are targeted for learning,
It is unrealistic because it requires a lot of computing power.

Further, in a window system capable of simultaneously starting a plurality of application programs, it is impossible to construct a system whose default value changes depending on the states of other independent application programs. .

Up to this point, the difficulty of using the default values among the application programs between users and the difficulty of not being able to set the default values among a plurality of application programs have been described. The same can be said for the difference in method.

That is, the method of operating the application program depends on the knowledge and experience of the user. For example, if you are using a document creation software to create a document, when moving the cursor, those who are accustomed to creating documents with a commercially available dedicated document creation machine can move the cursor with the arrow keys. Key bindings, for example, people who are used to writing documents with the emacs editor, which is widely used on workstations, tend to prefer to move the cursor using control (function) keys. There is a difference in (key allocation). In addition to this, some people prefer to move frequently used functions to the menu on the left side of the screen or keep them displayed as a palette.

In order to deal with such a difference in personal preference, it has become common for recent application programs to have a customization function. For example, the above-mentioned emacs provides a language for the user to freely change not only the key binding but also any operation, and the user can freely change it according to his or her preference. On the other hand, also on the user interface side, as described in "Application construction model for customization; Information Processing Society of Japan, Software Engineering Laboratory, 100-4, 1994.9.8", the user separates the function from the operation. Methods have been proposed for providing an environment in which the interface can be customized.

However, these conventional user interfaces are difficult to use because it is necessary for the user to set the environment for each application program every time it is installed and to adjust the operating method to suit the user's preference. There was a problem.

[0013]

As described above, in the conventional user interface, by predicting the state that the user will use, for example, the default value in each application program is automatically set, or the user's user interface is set. There is a problem that it is not easy to use because it is not possible to automatically set the operation method to suit your taste.

Therefore, an object of the present invention is to provide a method for customizing a user interface and an application program that can solve at least one of the above-mentioned problems and contribute to the improvement of usability.

[0015]

To achieve the above object, in the user interface according to the first aspect of the present invention, among the objects of each application program displayed on the display screen, the object designated in advance by the user is selected. First storage means for storing a correspondence table between the input destination and the output destination for each user, user determination means for determining the user who has started the application program, and this user determination means. Secondly storing an input example of the application program currently operated by the user to the specified input destination object
Storage means, the input case stored in the second storage means, and the correspondence table for the same user stored in the first storage means, and The default value of the object is inferred, the output destination object is set to the default value obtained by this inference, and the default value of the output destination object is forcibly set to a default value different from the inference result by the user. And default value inference means for learning the above inference rule.

Further, in order to achieve the above object, in the user interface according to the second aspect of the present invention, a customized information storing means for storing a customized operating method of each already installed application program, and this customized information storage. A rule extracting means for extracting a rule relating to customization from the customization information stored in the means; a reading means for reading rule and command information relating to the customization from a medium storing a newly installed application program; Customize the operation method of the newly installed application program by using the read rule and command information regarding customization and the rule extracted by the rule extraction means. And an inference means that.

In order to achieve the above object, a third aspect of the present invention is a method for customizing an operation method of each application program installed in a user interface according to a user, which is already installed in the user interface. A customized operation method of each application program is stored in a storage device as customization information that is expressed by a plurality of attributes and attribute values, and when the application program is newly installed, the user-specific customization is performed from the customization information. A rule regarding is extracted, and the operation method of the newly installed application program is customized based on this rule.

[0018]

In the user interface according to the first aspect of the present invention, the input case input by the user and the correspondence table for the same user stored in the first storage means are used and the output destination is set according to the inference rule. The default value of the object is inferred, and the output destination object is set to the default value obtained by this inference. Then, the inference rule is learned when the user changes the default value of the output destination object to a default value different from the inference result.

Therefore, every time the user starts the application program, the default value of the output destination object is automatically set to the initially set value. Then, when the user changes the default value of the output destination object to a default value different from the inference result, the inference rule is learned, and thereafter, the changed default value is automatically set. Can be switched.

Normally, the programs operating on the window system are Motif, Open Windows, Visual Contents.
It is created using a window program development environment such as. These development environments provide a library of objects required for displaying windows, such as initialization, buttons, menus, and input fields.

In the user interface according to the first aspect of the present invention, the objects provided in the window program development environment communicate with the default value inference means in a predetermined method to set the default values from outside the application program. That is, the window initialization function provided by the window program development environment (for example, X Tool Kit, which is the development environment of the X Windows System,
Xtinitialise function, X Vi, the development environment for Open Windows
ew is Xv (Init function) establishes communication with the default value inference means, and sets a default value while communicating with the default value inference means using the established communication path.

Many application programs operated by the GUI are created by combining objects such as menus, buttons, and input fields supplied as a library such as motif.
Therefore, the activation and communication of the default value inference means can be realized as the functions of these objects, and the application program can be created by the same programming method as the conventional one. Furthermore, an application program created in the past can be created by recompiling using a new library without changing the source code, and an application program corresponding to the default value inference means can be created.

In the user interface according to the second aspect of the invention and the application program customizing method according to the third aspect of the invention, for example, an attribute and an attribute value expressing the function or characteristic of the command are added to each command of each application program. A rule relating to the customization of the user is extracted from the customization information of the already installed application program, and the customization related to the operation of the newly installed application program is performed using this rule.

The attributes and attribute values added to the command of the application program are, for example, operation types: file operation, mode conversion, editing, cursor movement, ... Expected frequency of use: 10, 9, 8, 7, 6, 5, 4, 3,
2,1 Importance of operation: 10, 9, 8, 7, 7, 6, 5, 4, 3,
Refers to information such as 2, 1.

These attributes and attribute values are added to all customizable application programs. By doing this, it is possible to extract from the customization information that the user is currently using, for example, the rule that commands that are expected to be used frequently are placed in the leftmost menu, and also predict the application programs to be newly installed. It is possible to place frequently used commands in the leftmost menu, and customization can be performed automatically.

Further, the application program to be newly installed has a relation that should not be changed at the time of customization as a rule, and by using this and the extracted customization information for each user, consistency Achieved customization.

As an inference mechanism for extracting a rule regarding customization of a user from customization information and customizing based on this rule, Japanese Patent Application No.
05-113411 and C4.5 [C4.5: PROGRAMS FOR M
ACHIN LEARNING.J.Ross Quinlan], it is possible to use an inference device that extracts a rule from a case for an input inference condition and obtains a solution based on the rule.

[0028]

Embodiments will be described below with reference to the drawings. FIG. 1 is a block diagram of a user interface according to an embodiment of the present invention, which is a graphical user interface.

This embodiment is an example in which the present invention is applied to a window system capable of starting a plurality of application programs at the same time, and default values can be automatically set depending on the states of other independent application programs.

In the figure, reference numeral 1 denotes a processing device in which a plurality of application programs are installed, and 2
Is a display device for presenting information from the application program to the user, and 3 is for the application program by the user interactively selecting menus and buttons displayed on the screen of the display device 2 and inputting numerical values. The input device which operates is shown. Input device 3
The command input at is passed to the application program for processing.

The graphical user interface according to this embodiment includes a default value inference device P in addition to the above basic structure. The default value inference device P is roughly divided into
A user determination device 4 that determines the user who started the application program from the ID number and the like, and a default value in each application program that predicts the state that the user determined by this user determination device 4 will use, That is, the communication unit 6, which secures communication between the inference device 5 that automatically sets the default value and the inference device 5 and each application program installed in the processing device 6.
7 and 7. Note that one of the communication units can be omitted.

The inference apparatus 5 includes a control unit 8, a learning unit 9, and
The inference unit 10 and the storage unit 11 are included. In the storage unit 11, a correspondence table of the input destination and the output destination for the object designated by the user in advance among the objects of each application program displayed on the display screen is registered for each user.

When the user is identified by the user determination device 4, the control unit 8 receives the type of the event caused by the operation of the user via the communication units 6 and 7, and monitors the content of the event.
If the input example is for the input destination object listed in the correspondence table between the input destination and the output destination described above,
This value is stored. If the input case is an input case to the input destination object, the stored values of all the input destination objects are read out and passed to the inference unit 10.

The inference unit 10 reads inference rules from the storage unit 11 and infers a default value. The default value generated by inference is passed from the inference unit 10 to the control unit 8. The control unit 8 gives this default value to the designated output destination object of each application program via the communication units 7 and 6.

When there is an input to the output destination object due to the change setting by the user, the control unit 8 gives the value of the input destination object and the value of the output destination object to the learning unit 9. The learning unit 9 obtains these values and learns the inference rules stored in the storage unit 11.

Next, referring to FIG. 2, the flow of processing when the application program is activated will be described.
When an application program (probably an application program name is A1) is started in the processing device 1 (step 21), default value inference started by the same user who is waiting for the start of the application program A1 on the system. It is checked whether the device P exists (step 22). If it does not exist, the default value inference device P is activated (step 23). Then, communication is established with the default value inference device P (step 24) and the initialization is completed.

Next, the flow of processing of the control unit 8 at the time of startup will be described with reference to FIG. When the default value inference device P is activated, the control unit 8 obtains the activated application program name from the processing device 1 (step 31). Next, the setting file as shown in FIG. 4, which is stored in the storage unit 11 for each user, that is, the correspondence table of the input destination and the output destination for the object designated in advance by the user is read (step 32). The settings for the started application program (A1 in this case) are set (step 33). When communication with all the necessary application programs is established (step 34), the initial setting of the inference apparatus 5 is finished and inference is started.

If communication with all the necessary application programs has not been established (step 34), wait until the necessary application program is activated (step 35), and then establish communication with this application program. Perform (step 36).

In the setting file shown in FIG. 4,
The "required application program" column is used to register the application program name that must be activated for inference. In FIG. 4, the first entry shows that three application programs A1, A2, and A3 are required.
The "reasoning mechanism" column is used to specify a reasoning mechanism suitable for the nature of learning. The "inference rule" column is used to specify a file in which rules given to the inference mechanism and learning results up to now are recorded. The “input destination object” field is used to specify the object that is the source of inference. In addition, A1.m
enu 1 means menu 1 of application program A1
Means to depend on. The “output destination object” field is used to specify an object that gives a value obtained as a result of inference as a default value.

Next, the flow of the default value inference processing will be described with reference to FIG. First, the control unit 8 obtains the current value from the application program and initializes a variable that stores the current value of the input destination object (step 4).
1). If there is no value in the input destination object, no value (nil) is set as the value of the input destination object.

Next, the control unit 8 obtains an event from the application program (step 42). The event shall take the form of (application name: object name: event content). For example, when the first item is selected from the menu M1 of the application program A1, an event such as A1: M1: select1 is obtained.

When the received event is from the object designated by the "input destination object" in the setting file (step 43), it is recorded as the value of the input destination object (step 44). If the value of the corresponding input destination object already exists, it is overwritten.

The value is passed to the inference mechanism for inference (step 45), and the inference result is passed to the output destination object (step 46). The received event is not from the object specified in the "input destination object" of the configuration file (step 43), but the received event is from the object specified in the "output destination object" of the configuration file. If there is (step 47), the value of the input destination object and the value of the output destination object are given to the learning unit 9 to perform learning (step 48), and the process returns to step 42 to wait for the next event.

When the received event is not from the object designated by the "output destination object" of the setting file (step 47) and when the received event is the application end event (step 49) Ends the process.

If the received event is not the application end event (step 49), the process returns to step 42 to wait for the next event. Here, the flow of processing will be described using a specific example.

For example, assume that the activated application program 1 has a menu as shown in FIG. This menu is referred to as menu 1, and the buttons of the menu are referred to as button 1a, button 1b, and button 1c. It is also assumed that the started application program 2 has a menu as shown in FIG. This menu will be referred to as menu 2, and each button of the menu will be referred to as button 2a, button 2b, and button 2c.

When the application program 1 is started, the user is determined (here, user A), the default value inference device P started by the user A is searched, and this is waiting for the start of the application program 1. In some cases, establish communication. In this case, since there is no default value inference device P waiting for the activation of the application program 1, the default value inference device P is activated to establish communication.

The activated default value inference device P reads a setting file as shown in FIG. 8, that is, a correspondence table,
After completing the initial setting, the application program 2 is waited for.

When the application program 2 is started up, the user is judged (here, user A), the default value inference device P started up by the user A is searched, and this waits for the application program 2 to start up. In some cases, establish communication. In this case, there is a default value inference device P waiting for the activation of the application program 2, so the default value inference device P is activated to establish communication.

Next, it is assumed that the user selects the button 1a of the menu 1 in the application program 1. The application program 1 notifies the default value inference device P via the communication unit that the button 1a of the menu 1 has been selected. Since the menu 1 is specified as the input destination object in the setting file shown in FIG. 8, the button 1a having this value is passed to the inference mechanism specified in the setting file. Since the neural network is specified here, the neural network is activated and the inference rule is set to the state of the network here. If the value obtained by the neural network is the button 2a, the default value of the menu 2 in the application program 2 is set to the button 2a via the communication unit.

Next, when the user selects the button 2b of the menu 2 in the application program 2, the default value inference device P pushes the button for the input of the value of the menu 1 of the application program 1 (currently the button 1a). Train the neural net to get 2b as the answer.

As can be seen from the above operation, the state / value that the user will use can be predicted and automatically set as the default state / value without the user performing complicated programming. The above states / values can be automatically set without being aware of them at the stage of creating the application program. Therefore, the work efficiency in each application program can be improved.

It should be noted that, even if the inference mechanism is a case-based inference method or a method of inferring from a simple frequency or a combination that has occurred most recently, the default value can be obtained as in the above embodiment.

FIG. 9 is a block diagram of the essential parts of a user interface according to another embodiment of the present invention. It should be noted that this embodiment is an example in which, when a new application program is installed, the present invention is applied so that the operation method can be automatically customized so as not to cause inconsistency with the existing application program.

In the figure, reference numeral 61 denotes an input device operated by the user. When the application device installation start command is given from the input device 61 to the control device 62, the control device 62 notifies the rule extraction unit 63 of the rule extraction start.

Upon receiving the rule extraction start notification, the rule extraction unit 63 reads out the customization information of the already installed application program from the customization information storage unit 64, and extracts the rule regarding the customization of the user based on this information. Then, the extracted rule is passed to the inference unit 65.

The inference unit 65 which has received the rule reads the rule when customizing the new application program to be installed, and reads it in the device 66.
Read from the application storage medium 67 via.

The inference unit 65 matches the read customization rule with the rule previously extracted by the rule extraction unit 63. When a contradiction occurs, the customization rule is prioritized and the extracted rule is extracted. Discard the conflicting ones. Further, the inference unit 65 reads the information about the command of the new application program to be installed by the reading device 66.
Read from the application storage medium 67 via
Inference is performed using information about rules and commands, and the result of inference is stored in the customized information storage unit 64.

Next, the flow of processing in the rule extraction unit 63 will be described with reference to FIG. When the rule extraction start command is given from the control device 62, the rule extraction unit 63
First, various variables for storing rules and the like are initialized (step 101).

Next, one application program stored in the customization information storage unit 64 is searched (step 102). If there is an unprocessed application program (step 103), the customization information is read (step 102). 104).

When all the customization information has been read (step 103), rules are extracted from this (step 105), and the extracted rules are passed to the inference unit 65 to end (step 106).

Here, the customization information storage unit 64 is
It is assumed that the information as shown in FIG. 10 is stored. FIG. 10 shows a part of the customization information of the document creation software. In addition to this, for example, spreadsheet software is also stored, and the customization information has contents as shown in FIG. Each attribute in each figure has the following meaning.

Function name: A label for the user to identify, not used for rule extraction. Target: Indicates the main target for the operation. Action: Indicates the action performed by the operation.

Direction: Indicates the direction of movement. Prediction frequency: This is a standard usage method and indicates how often the function is used in 10 levels (the higher the frequency, the larger the value).

Importance: Shows how important the function is in a standard use method in 10 levels (the more important the value, the larger the value). Operation: Indicates the operation performed when using the function. ctrl represents the operation performed while pressing the control key (function key). ctrl-n represents pressing the letter "n" while holding down the control key. menu-1-1 indicates that it is the first item in the first menu. pallete-
1-1 indicates that it is the first item in the first menu palette.

User setting: The case of ◯ means that the user manually customized, not the automatic customization. Note that the menu displays a list of menu items only during the operation of selecting a menu, whereas the menu palette always displays a list of menu items as one independent window.

When such customization information is given, the rule extraction unit 63 extracts the following rules. For example, cursor movement, deletion of one line, file saving, etc., which are operated in the same way, generate rules with the same customization, for example, as a rule for cursor right movement, cursor and movement and right and10 and10 → ctrl- produces f.

Also, despite the same command,
In the case of the table mode, which uses different operation methods, create documents: table mode, mode, switching, table, 2, 2, menu
-3-3 Spreadsheet: table mode, mode, switching, table, 6, 8, pallet
From e-1-3, the difference is extracted and used as a rule. For example, in this case, since the prediction frequency and the importance are different, the prediction frequency> 4 and the importance> 6 → pllete-1-3 prediction frequency ≦ 4 and the importance ≦ 6 → menu-3-3. It should be noted that the thresholds 4 and 6 simply take intermediate values between the prediction frequency and the importance. However, the prediction frequency ≤ 4 and the importance ≤ 6 → The rule of menu-3-3 contradicts the rule of cursor movement, so the mode and switching and the prediction frequency <4 and the importance <6 →
It becomes like menu-3-3. Also, character mode, mode, switching, character, 7, 7, pallete-1-
1 figure mode, mode, switching, figure, 7, 7, pallete-1-
2 Table mode, mode, switching, table, 2, 2, menu-3-3 From table mode, mode, switching, table, 6, 8, pallete-1-3, mode and switching and prediction frequency> 4 and importance > 4.5 →
A rule is generated that a menu palette is used for important mode switching commands that have a high prediction frequency such as pallte-1.

However, the items manually set by the user in the user setting are not used for the rule extraction, and the case is directly used as the rule. The generated rule is passed to the inference unit 65 in the form shown in FIG. The attribute value "-"
Indicates that there is no particular designation. The other attributes with embedded attribute values mean that the operation is determined by the operation if the combination of the attribute and the attribute value is satisfied. Further,> 4 in the numerical value column means that the attribute value of the corresponding attribute is larger than 4. If only a numerical value is entered, it means that the condition is satisfied only when the numerical value is equal. Furthermore, if the end of the value of the operation ends with "-" like "ctrl-",
Indicates that some value other than "ctrl-" is required.

Next, the flow of processing in the inference unit 65 will be described with reference to FIG. First, the variables to be used are initialized (step 201), and the rule extraction unit 63 is
The rule list thus extracted is received and stored in a variable (step 202). When there is an unprocessed rule specific to the application stored in the application storage medium 67 (step 203), one rule is read (step 204). The consistency between the read rule and the rule received from the rule extraction unit 63 is checked. If there is no contradiction (step 205), the read rule is added (step 207), and if there is a contradiction (step 205). ), Delete all inconsistent rules (step 206), add the read rules (step 207), and return to step 203 to check if there are any unprocessed rules.

If there is no unprocessed rule (step 203), the command is customized. If there is unprocessed command information of the application program stored in the application storage medium 67 (step 208), the command is read (step 209), the operation to be assigned is obtained from the rule (step 210), and step 208. Then, it is checked whether or not there is an unprocessed command on the application storage medium 67.

When there is no unprocessed command (step 208), the customization information of the new application program is written in the customization information storage section 64 (step 211), and the process is ended.

Next, the method for detecting contradiction in step 205 will be described. Application storage medium 67
The rules above are written the same as the extracted rules. For example, suppose there is the following rule.

Mode, switching,>4,> 4.5,-, pallte-2- This means that the command relating to the mode switching is the menu palette 2. This is different from one of the extracted rules, that is, mode, switching,>4,> 4.5,-, and pallete-1-, even though both have the same condition. In such a case, the extracted rule is deleted.

Next, a method for obtaining an operation to be assigned to a command in step 210 will be described. The command information on the application storage medium 67 is described as cursor right move: cursor, move, right, 10, 10 and right arrow key, like the customization information. The last right arrow key is the operation method recommended by the application program. This operation method is used when there is no need to change it by customization.

If a rule that matches the conditions is applied to the above command information, cursor and movement and expected frequency = 10 and importance = 10
→ ctrl- Right → f Two rules match. Therefore, the cursor right movement becomes ctrl-f.

Similarly, if word deletion: word, delete, none, 5, 5, menu-3-3 exists as a command and the following rules exist, word,-,-,-,-, ESC- -, Delete,-,-,-, k ESC- k is registered as a word deletion.

If there are the following rules, there is a situation in which line,-,-,-,-, ESC- line, move,-,-,-, ctrl- can be used. . In this case, the one with the larger number of matching attributes is selected.

As described above, from the customization information of the already installed application program,
Since the user's rules regarding customization are extracted and the application program to be newly installed is customized using this rule,
It is possible to automatically form a consistent operating environment without the user performing complicated customization work. Also, because it can provide the operating environment that suits the user's taste,
The time required to learn the operation of a new application program can be shortened, which contributes to the efficiency of work.

[0080]

As described above, according to the present invention,
Since the default value of each application program can be automatically set or the customization of each application program can be automated according to the user, it is possible to contribute to the improvement of usability.

[Brief description of drawings]

FIG. 1 is a block diagram of a user interface according to an embodiment of the present invention.

FIG. 2 is a diagram for explaining the flow of processing in the interface.

FIG. 3 is a diagram for explaining the flow of processing in the interface.

FIG. 4 is a diagram showing an example of a correspondence table used in the interface.

FIG. 5 is a diagram for explaining the flow of processing in the interface.

FIG. 6 is a diagram for explaining a specific processing example in the interface.

FIG. 7 is a diagram for explaining a specific processing example in the interface.

FIG. 8 is a diagram showing an example of a correspondence table used in the same example.

FIG. 9 is a block diagram of a main part of a user interface according to another embodiment of the present invention.

FIG. 10 is a diagram showing an example of customization information stored in a customization information storage unit of the interface.

FIG. 11 is a diagram showing another example of the customization information stored in the customization information storage unit of the interface.

FIG. 12 is a diagram showing an example of rules extracted from the customization information stored in the customization information storage unit.

FIG. 13 is a diagram for explaining the processing flow of the interface.

FIG. 14 is a diagram for explaining the processing flow of the interface.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Processing device 2 ... Display device 3 ... Input device 4 ... User determination device 5 ... Inference device 6, 7 ... Communication part 8 ... Control part 9 ... Learning part 10 ... Inference part 11 ... Storage part 61 ... Input device 62 ... Control device 63 ... Rule extraction unit 64 ... Customized information storage unit 65 ... Inference unit 66 ... Reading device 67 ... Application storage medium P ... Default value inference device

Claims (3)

[Claims] 1. A user interface for interactively executing an operation for each application program via a display screen, wherein an object specified in advance by a user among objects of each application program displayed on the display screen is displayed. First storage means for storing a correspondence table of input destinations and output destinations for each user, user determination means for determining the user who has started the application program, and usage determined by the user determination means Second storage means for storing an input case of the application program currently being operated by the person to the specified input destination object, and the input case and the first storage case stored in the second storage means. Using the correspondence table for the same user stored in the storage means, the inference rule is The default value of the output destination object is inferred according to the rule, and the default value of the output destination object is set as the inference result by the user while setting the output destination object to the default value obtained by this inference. A user interface comprising a default value inference means for learning the above inference rule when the default value is changed to a different default value. 2. A user interface in which a plurality of application programs can be installed, and customization information storage means for storing a customized operation method of each already installed application program, and the customization information storage means. A rule extracting means for extracting a customization rule from the customization information, a reading means for reading the customization rule and command information from a medium storing a newly installed application program, and a customization rule read by the reading means. And operating method of application program newly installed using command information and rules extracted by the rule extraction means A user interface comprising an inference means for customizing a law. 3. A method of customizing an operation method of each application program installed in a user interface according to a user, wherein a plurality of customized operation methods of each application program already installed in the user interface are provided. It is stored in a storage device as customization information expressed by attributes and attribute values, and when a new application program is installed, a rule relating to the customization unique to the user is extracted from the customization information, and a new rule is created based on this rule. A method of customizing an application program, characterized in that the operation method of the application program installed in is customized.