JP6428459B2 - Operation determination program, operation determination method, and operation determination device - Google Patents

Description

  The present invention relates to an operation determination program, an operation determination method, and an operation determination apparatus.

  Conventionally, applications that are frequently started are identified based on the application startup history, and the top screen that is displayed first when the device is started and the application icons that are frequently started on the launcher are automatically displayed. There is technology to arrange. As a result, it is easy to perform an activation operation of an application that is frequently activated.

  Further, after the application is started, if the application is terminated within a time period in which it can be assumed that the application is not used, the application may not be started and may not be recorded in the application activation history. Accordingly, when an application that is frequently activated is identified from the activation history of the application, it is possible to prevent erroneously identifying an application that is frequently activated based on an unintentionally activated history.

  As prior art, for example, it is judged whether or not the driver intends to operate the brake by comparing the area ratio of the portion where the driver's foot is placed with a threshold value determined by statistics. There is technology. In addition, for example, there is a technique for taking statistical values such as an average, minimum value, and mode value of the intervals at which a button is pressed, and increasing the error probability of the button when the statistical value is equal to or less than a predetermined threshold value.

JP 2008-063953 A JP 2012-220994 A

  However, with the conventional technology, it is difficult to reduce the fact that an application activation instruction is an intended operation but is not an unintended operation. For example, if a minimum value or an average value is used as a threshold value for the time when it is assumed that the application is not used, an unintended operation may be determined as an intended operation. On the other hand, if the threshold value is increased in order to reliably remove unintended operations, the intended operation is determined to be an unintended operation.

  In one aspect, an object of the present invention is to provide an operation determination program, an operation determination method, and an operation determination device that reduce an application start instruction that is determined as an unintended operation although it is an intended operation. To do.

According to one aspect of the present invention, when determining whether the activation instruction of the application is an intended operation or an unintended operation based on whether the execution time from the activation of the application to the termination is equal to or greater than a threshold value, Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time, and the time at the point corresponding to the bottom of the peak including the predetermined time among the peaks of the frequency distribution of the generated execution time An operation determination program, an operation determination method, and an operation determination device for determining a threshold value are proposed.

  According to an aspect of the present invention, there is an effect that it is possible to reduce the determination that an application activation instruction is an intended operation but is not intended.

FIG. 1 is an explanatory diagram of an example of the operation determination method according to the embodiment. FIG. 2 is a block diagram illustrating a hardware configuration example of the operation determination apparatus 101. FIG. 3 is an explanatory diagram showing an example of the operation history DB 300. FIG. 4 is an explanatory diagram showing an example of the execution time recording table 400. FIG. 5 is an explanatory diagram showing an example of the execution time recording table 500 for each application. FIG. 6 is an explanatory diagram showing an example of the execution time frequency table 600. FIG. 7 is a block diagram illustrating a functional configuration example of the operation determination apparatus 101. FIG. 8 is a flowchart illustrating an example of a threshold determination process procedure of the operation determination apparatus 101. FIG. 9 is a flowchart illustrating an example of the upper limit calculation process procedure of the operation determination apparatus 101. FIG. 10 is an explanatory diagram illustrating a first specific example of threshold determination of the operation determination device 101. FIG. 11 is an explanatory diagram illustrating a second specific example of threshold determination performed by the operation determination device 101.

  Hereinafter, embodiments of an operation determination program, an operation determination method, and an operation determination apparatus according to the present invention will be described in detail with reference to the drawings.

(Embodiment)
FIG. 1 is an explanatory diagram of an example of the operation determination method according to the embodiment. The operation determination apparatus 101 in FIG. 1 determines a threshold value for each application program (hereinafter abbreviated as an application), and compares the execution time of the application with the threshold value to determine whether the application activation instruction is the intended operation. It is a device that determines whether or not. Specifically, the operation determination apparatus 101 is a computer such as a PC (personal computer) or a tablet terminal.

  Here, the execution time is the time from the start of the application to the stop. The activation instruction is an instruction to activate an application used by the user. For example, the activation instruction is tapping an icon for activating an application or double-clicking with a mouse. The stop means that the user ends the use of the application. For example, the stop includes terminating the use of the application by switching to the application. Further, the stop includes terminating a task for executing the application.

  The execution time is the time from when the application starts up until the application stops. The intended operation is a normal operation for starting an application that the user is trying to start. An unintended operation is an erroneous operation that activates another application instead of the application that the user tried to activate.

  Conventionally, a smartphone has a function of presenting an application that a user is likely to use in order to improve user operability. For this reason, it calculates which application is likely to be activated from the execution history of the activated application, and presents an application that is likely to be used based on the result. Here, the execution history is a history for recording the time when the application is started and the time when it is stopped.

At this time, since the user misoperates the device due to various causes, operations unintended by the user are also stored in the application execution history. For example, in the case of a smartphone, there are the following reasons for accidentally pressing a button for starting an application.
・ I touched the button even though I didn't intend to press it.
・ Your finger did not hit the correct position.
・ It overlaps with the timing when the screen changes, resulting in pressing an unintended button.
・ The button was pressed incorrectly.
・ You pressed a different button due to wrong or vague memory or knowledge.

  When an unintended operation is included in the execution history, an application activation probability is calculated by interpreting that an unintended operation is also desired by the user. In addition, an unintended operation may occur at a considerable rate. In this case, an application that is not intended by the user may be determined as an application that is frequently started. For example, when there is an application that a user often starts unintentionally on a smartphone, it may be determined that the application is often used.

  Therefore, a method of removing data due to an unintended operation when determining that the application is frequently started from the execution history can be considered. For example, when an application is activated by an unintended operation, the user immediately terminates the application, so the execution time of the application is shorter than when the application is activated by an intended operation. For this reason, a threshold value is provided, and if it is equal to or less than the threshold value, it can be determined that the data is an unintended operation. For example, this threshold value may be determined by a statistical method.

  However, when the minimum value or mode value determined by a statistical method is used, an unintended operation may be determined as an intended operation. For this reason, the threshold value can be increased in order to prevent an unintended operation from being determined as an intended operation. In this case, the number of intended operations decreases because the number of unintended operations increases despite the intended operation. If the number of intended operations decreases, the number of activations may be small even though the application is originally frequently activated, so that it may not be determined as an application that is frequently activated.

  Therefore, in the embodiment, the operation determination device 101 generates an execution time frequency distribution from the application execution time history, and corresponds to the bottom of a peak including a predetermined time among the peaks of the execution time frequency distribution. Time is determined as a threshold. Thereby, the operation determination apparatus 101 can make the operation included in the peak including the predetermined time the intended operation. For this reason, the operation determination apparatus 101 can reduce an application start instruction being determined as an unintended operation even though it is an intended operation.

  Hereinafter, an example of the operation determination apparatus 101 according to the embodiment will be described. In the example of FIG. 1, the threshold value of the application A is determined based on the history of the execution time of the application A. Hereinafter, application A may be abbreviated as application A.

  (1) The operation determination apparatus 101 generates a frequency distribution 111 for each application from the storage unit 110 that stores an execution time from when the application is started to when it is stopped. Here, for example, the storage unit 110 stores a history of application start times and application stop times. The operation determination apparatus 101 calculates the execution time of the application by subtracting the start time from the stop time, and generates the frequency distribution 111 by counting the number of start times of the application within a certain interval of the execution time.

  In the example of FIG. 1, the storage unit 110 stores a history of the start time of the application A and the stop time of the application A. The operation determination apparatus 101 generates the frequency distribution 111 of the application A from the storage unit 110.

  (2) The operation determination apparatus 101 extracts a peak including a predetermined time from the peaks of the frequency distribution 111. Here, the predetermined time is a predetermined value and can be determined in advance by a statistical method or the like. For example, the predetermined time can be obtained from the probability that the user makes an error in starting the application. For example, the operation determination apparatus 101 can set the predetermined time as a time in which the ratio of the number of data shorter than the predetermined time to the total number of data is lower than the probability that the user makes an error in starting the application. Here, the peak is a mountain having a maximum value until the frequency increases, decreases and then increases in the frequency distribution.

  In the example of FIG. 1, the operation determination apparatus 101 detects three peaks 120, a peak 130, and a peak 140, and extracts a peak 140 that includes a predetermined time.

  (3) The operation determining apparatus 101 determines the time corresponding to the bottom of the extracted peak 140 as a threshold for determining whether the application activation instruction is an intended operation or an unintended operation. Here, the tail is a point where the frequency on the side where the extraction time of the extracted peak is small starts to decrease and increases.

  In the example of FIG. 1, the operation determination apparatus 101 determines a point 150 whose frequency starts to increase at the peak 140 as a threshold for determining whether the application activation instruction is an intended operation or an unintended operation.

  As described above, the operation determination apparatus 101 generates the execution time frequency distribution 111 with reference to the storage unit that stores the application execution time history. After that, the operation determination apparatus 101 determines the time at the point corresponding to the bottom of the peak including the predetermined time among the peaks of the generated frequency distribution 111 of the execution time as the threshold value. Thereafter, the operation determination apparatus 101 determines whether the application activation instruction is an intended operation or an unintended operation based on whether or not the execution time from when the application is activated until it is stopped is equal to or greater than a threshold value.

  Thereby, the operation determination apparatus 101 can change the operation included in the peak including the predetermined execution time to the intended operation. This is due to the following reason. For example, when the operation determination apparatus 101 generates a frequency distribution of execution times, a plurality of peaks are detected. In either case of an intended operation or an unintended operation, there is a trend in the length of execution time. For this reason, the operations included in these peaks are either intended or unintended operations. Here, in order to prevent an unintended operation from being determined as an intended operation, the peak including the predetermined time is a peak including the intended operation when the predetermined time is a large value. In this case, when this peak is divided at a predetermined time, the tail part of the peak is determined to be an unintended operation despite the intended operation. For this reason, the operation determination apparatus 101 can reduce the determination that the intended operation is an unintended operation by setting the determination threshold to the bottom of the peak. Therefore, in order to eliminate unintended operations, even if the predetermined time is increased, the number of intended operations does not decrease, so that the operation determination apparatus 101 reduces errors in determining frequently started applications. can do.

(Hardware configuration example of operation determination apparatus 101)
FIG. 2 is a block diagram illustrating a hardware configuration example of the operation determination apparatus 101. In FIG. 2, the operation determination apparatus 101 includes a CPU (Central Processing Unit) 201, a memory 202, an I / F (Interface) 203, a disk drive 204, a disk 205, a display 206, an input device 207, Have Each component is connected by a bus 200.

  Here, the CPU 201 governs overall control of the operation determination apparatus 101. The memory 202 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and a flash ROM. Specifically, for example, a flash ROM or ROM stores various programs, and the RAM is used as a work area of the CPU 201 as a storage unit. The program stored in the memory 202 is loaded on the CPU 201 to cause the CPU 201 to execute the coded process.

  The I / F 203 is connected to a network 210 such as a LAN, WAN, or Internet through a communication line, and is connected to another computer via the network 210. The I / F 203 controls an internal interface with the network 210 and controls input / output of data from other computers. For example, a modem or a LAN adapter may be employed as the I / F 203.

  The disk drive 204 controls reading / writing of data with respect to the disk 205 according to the control of the CPU 201. The disk 205 stores data written under the control of the disk drive 204. Examples of the disk 205 include a magnetic disk and an optical disk.

  A display 206 displays data such as a document, an image, and function information as well as a cursor, an icon, or a tool box. As the display 206, for example, a liquid crystal display, an organic EL (Electroluminescence) display, or the like can be adopted.

  The input device 207 has keys for inputting characters, numbers, various instructions, and the like, and inputs data. The input device 207 may be a keyboard or a mouse, or may be a touch panel type input pad or a numeric keypad.

  Note that the operation determination apparatus 101 may not include, for example, the disk drive 204 and the disk 205 among the above-described components. The operation determination apparatus 101 may include, for example, an SSD (Solid State Drive), a scanner, and a printer in addition to the above-described components.

(An example of the operation history DB 300)
FIG. 3 is an explanatory diagram showing an example of the operation history DB (database) 300. In FIG. 3, for example, the operation history DB 300 is created and updated by the operation determination device 101 and stored in the memory 202 or the disk 205 of the operation determination device 101.

  The operation history DB 300 is a DB that stores events that have occurred in the operation determination apparatus 101. Here, the event is an operation performed by the operation determination device 101 by an operation from the user. For example, the event includes ON / OFF of the display 206 of the operation determination device 101 by the user, and activation of an application of the operation determination device 101 by the user. In the following description, an application may be simply referred to as an application.

  The operation history DB 300 has items of time and operation. The operation history DB 300 stores the operation history information as one record by setting information for each item. For example, the operation history DB 300 illustrated in FIG. 3 stores operation history information 300-1 to 300-n as each record.

  Here, the time indicates the time when the event occurred in the operation determination apparatus 101. For example, the time is expressed in hours, minutes, and seconds. The operation indicates the type of event that has occurred in the operation determination apparatus 101.

  In the example of FIG. 3, the record 300-1 indicates that the application A is activated at 10:10:20 in the operation determination apparatus 101.

(Example of execution time record table 400)
FIG. 4 is an explanatory diagram showing an example of the execution time recording table 400. In FIG. 4, the execution time recording table 400 is created and updated by, for example, the operation determination device 101 and stored in the memory 202 or the disk 205 of the operation determination device 101.

  The execution time record table 400 is obtained by extracting from the operation history DB 300 a record in which the operation item is an application activation or application termination.

  The execution time recording table 400 includes items of time and application start / stop. The execution time recording table 400 stores the execution time information as one record by setting information for each item. For example, the execution time record table 400 shown in FIG. 4 stores execution time information 400-1 to 400-6 as each record.

  Here, the time indicates the time when the activation or stop of the application occurs in the operation determination apparatus 101. For example, the time is expressed in hours, minutes, and seconds. In addition, the activation stop of the application indicates that the application is activated or stopped in the operation determination apparatus 101.

  In the example of FIG. 4, the record 400-1 is extracted from the record 300-1 of the operation history DB 300 and indicates that the application A is started at 10:10:20 in the operation determination device 101. .

(Example of execution time record table 500 for each application)
FIG. 5 is an explanatory diagram showing an example of the execution time recording table 500 for each application. In FIG. 5, an execution time recording table 500 for each application is created and updated by, for example, the operation determination apparatus 101 and stored in the memory 202 or the disk 205 of the operation determination apparatus 101.

  The execution time recording table 500 for each application is obtained by dividing the execution time recording table 400 for each application.

  The execution time record table 500 for each application includes items of time and start / stop of the application. The execution time record table 500 for each application stores the execution time information for each application as one record by setting information for each item. For example, the execution time record table 500 for each application shown in FIG. 5 stores execution time information 500-1 to 500-6 and 510-1 to 510-4 for each application as records.

  Here, the time indicates the time when the activation or stop of the application occurs in the operation determination apparatus 101. For example, the time is expressed in hours, minutes, and seconds. In addition, the activation stop of the application indicates that the application is activated or stopped in the operation determination apparatus 101.

  In the example of FIG. 5, records 500-1 to 500-6 are obtained by extracting the start and stop of the application A from the execution time recording table 400. Here, the records 500-3 to 500-6 are assumed to exist in the portion of “...” In the example of FIG. The same applies to the records 510-3 to 510-4.

(Example of execution time frequency table 600)
FIG. 6 is an explanatory diagram showing an example of the execution time frequency table 600. In FIG. 6, the execution time frequency table 600 is created and updated by, for example, the operation determination apparatus 101 and stored in the memory 202 or the disk 205 of the operation determination apparatus 101.

  The execution time frequency table 600 is a frequency distribution of execution times generated for each application from the execution time recording table 500 for each application.

  The execution time frequency table 600 includes items of execution time, data number, cumulative data number ratio, data number difference, and difference change. The execution time frequency table 600 stores the execution time frequency information as one record by setting information for each item. For example, the execution time frequency table 600 illustrated in FIG. 6 stores execution time frequency information 600-1 to 600-15 as each record.

  Here, the execution time indicates the time from the start of the application to the end in seconds. The execution time is set at a constant time interval. For example, the execution time can be set at 1 second intervals. Further, when n is set as the execution time, the time from when the application is activated until it is terminated is between n−1 seconds and n seconds.

  The number of data indicates the number of times the application has been started since it was started during the execution time. The cumulative data number ratio indicates the ratio of the number of times that the application has been started and ended after the time until the execution time to the total data number. The data number difference indicates a difference in the number of data between the record and the record immediately above. Here, the record immediately above is a record having an execution time shorter by a certain time than the execution time of the record. Also, the first record is obtained with the data number difference assuming that the data number of the record one above is 0.

  Further, the item of change in difference indicates whether or not the difference in the number of data is increased as compared with the record one above. Here, the operation determination device 101 sets “increase” when the data number difference increases, sets “decrease” when the data number difference decreases, and is blank when the data number difference does not change. Set. Further, the first record obtains a change in the difference on the assumption that the data number difference of the record one above is 0.

  The example of FIG. 6 shows the frequency distribution of the execution time of the application A. In the example of FIG. 6, the record 600-2 indicates that the number of times the application has been started from 1 second to 2 seconds and ended is 2 times, and the cumulative data number ratio is 6%. . In the record 600-2, since the number of times the application A is started from 0 second to 1 second is ended once, the data number difference is 1, and the data number difference is the upper record. Since they are the same, the change in the difference is blank.

(Functional configuration example of the operation determination device 101)
FIG. 7 is a block diagram illustrating a functional configuration example of the operation determination apparatus 101. In FIG. 7, the operation determination apparatus 101 includes a storage unit 701, a determination unit 702, and a determination unit 703. Specifically, the control unit including the determination unit 702 and the determination unit 703, for example, causes the CPU 201 to execute a program stored in a storage device such as the memory 202 illustrated in FIG. Realize. The processing result of each functional unit is stored in, for example, a storage device such as the memory 202 illustrated in FIG.

  The storage unit 701 has a function of storing events that have occurred in the operation determination apparatus 101. For example, when an event occurs in the operation determination device 101, the storage unit 701 stores the time when the event occurred and the type of the event in the operation history DB 300.

  In addition, the storage unit 701 has a function of storing an execution time recording table 400, an execution time recording table 500 for each application, and an execution time frequency table 600 that are generated for the determination unit 702 to determine a threshold value.

  The determination unit 702 has a function of calculating a threshold value for each application for determining whether the application activation instruction is an intended operation or an unintended operation.

  For example, the determination unit 702 generates the execution time recording table 400 from the operation history DB 300. The operation history DB 300 includes records of all events that have occurred in the operation determination apparatus 101. For this reason, the determination unit 702 generates an execution time recording table 400 by extracting application start / stop records from the operation history DB 300.

  For example, the determination unit 702 generates an execution time recording table 500 for each application from the execution time recording table 400. The execution time record table 400 includes records of all applications that have been operated by the operation determination apparatus 101. Therefore, the determination unit 702 generates an execution time recording table 500 for each application by extracting a record for each application from the execution time recording table 400.

  For example, the determination unit 702 generates an execution time frequency table 600 for each application from the execution time recording table 500 for each application. First, the determination unit 702 generates the number of data in the execution time frequency table 600. Next, the determination unit 702 calculates the cumulative data number ratio of the execution time frequency table 600. Next, the determination unit 702 calculates the data number difference in the execution time frequency table 600. Finally, the determination unit 702 sets a change in the difference in the execution time frequency table 600.

The determination unit 702 detects a peak from the execution time frequency table 600 where the change in the difference matches the pattern of “increase → decrease → increase”. For example, the operation determination apparatus 101 can detect the following peaks.
(I) For example, when the user activates the application, the peak when the application is immediately stopped and the application is stopped (ii) For example, when the user activates the application, the application screen is displayed. Peak when the application is stopped after being recognized as an unintended application (iii) For example, when the user starts the application, the user refers to the displayed information and then stops the application Peak (iv) For example, when the user starts the application, the user refers to the displayed information, and then inputs the information into the application and then stops the application

  Further, the determination unit 702 calculates the upper limit value of the execution time of the error operation from the execution time frequency table 600. For this reason, the determination unit 702 detects a record in which the cumulative data number ratio in the execution time frequency table 600 exceeds the error ratio. Here, the error rate is a value input to the operation determination device 101 by the user. For example, the user can input an error rate within a range of 0% to 100% as a probability of making an error in starting the application. The determining unit 702 can set the execution time of the detected record as the upper limit value of the execution time of the error operation. Here, the determination unit 702 detects a record in which the cumulative data number ratio exceeds the error ratio in order to detect many erroneous operations by the user.

  The determination unit 702 can extract a peak including the upper limit value of the execution time of the calculated error operation. For example, the determination unit 702 extracts the first record whose difference change is “increase” from the extracted peaks. The determination unit 702 can set the execution time of the extracted record as a threshold value.

  The determining unit 702 can also extract the maximum peak that does not exceed the upper limit value of the calculated error operation execution time. Here, the maximum peak is a peak having the maximum execution time at the peak apex. For example, the determining unit 702 extracts the first “increase” record after the change in the difference becomes “decrease” in the extracted peaks. The determination unit 702 can set the execution time of the extracted record as a threshold value.

  Here, the determination unit 702 may not be able to extract a peak that does not exceed the upper limit of the execution time. In addition, the determination unit 702 may not be able to extract the first “increase” record after the change in the difference becomes “decrease” in the extracted peaks. In these cases, the determination unit 702 can be interpreted as an application that is not activated by an error operation. For this reason, the determination unit 702 can set the threshold to zero.

  The determination unit 703 determines whether the application activation instruction is an intended operation or an unintended operation using a threshold value calculated for each application. For example, when the determination unit 703 is shorter than the threshold value calculated for each application, the determination unit 703 determines that the application activation instruction is an unintended operation. In addition, when the determination unit 703 is equal to or greater than the threshold calculated for each application, the determination unit 703 determines that the application activation instruction is the intended operation.

(Example of threshold determination processing procedure)
FIG. 8 is a flowchart illustrating an example of a threshold determination process procedure of the operation determination apparatus 101. In FIG. 8, first, the operation determining apparatus 101 generates an execution time recording table 400 from the operation history DB 300 (step S801). For example, the operation determination apparatus 101 generates an execution time recording table 400 by extracting a record of application start and application stop from the operation history DB 300.

  For example, in the example of FIG. 3, the operation determination apparatus 101 records records in which the operation items in the operation history DB 300 are application A activation, application A stop, application B start, application B stop, application C start, and application C stop. Extract.

  Next, the operation determination apparatus 101 generates an execution time recording table 500 for each application from the execution time recording table 400 (step S802).

  For example, the operation determination apparatus 101 in the example of FIG. 5 is an example of the execution time recording table 500 of the application A and the execution time recording table 500 of the application B. The operation determination apparatus 101 also generates an execution time recording table 500 for the application C, which is not illustrated in FIG.

  Next, the operation determination apparatus 101 generates an execution time frequency table 600 for each application from the execution time recording table 500 for each application (step S803).

  For example, the operation determination apparatus 101 generates the number of data in the execution time frequency table 600. Specifically, the operation determination apparatus 101 extracts an application activation record from the execution time recording table 500 for each application, and extracts an application stop record at a record lower than the extracted record. Here, the lower record is a record having an execution time longer than the execution time of the record. In the execution time frequency table 600, the operation determination apparatus 101 adds 1 to the number of execution time data corresponding to the time difference between the extracted records. The operation determination apparatus 101 performs a process of adding the number of data in the execution time frequency table 600 to all the records in the execution time record table 500 for each application.

  Next, the operation determination apparatus 101 calculates the cumulative data number ratio in the execution time frequency table 600. For example, the operation determination apparatus 101 calculates the cumulative data number ratio by adding the data numbers of records above the record for which the cumulative data number ratio is calculated and dividing the added value by the total data number. Here, the upper record is a record having an execution time shorter than the execution time of the record.

  In the example of FIG. 6, the number of data when the execution time is 0 to 1 second is 1, the number of data when the execution time is 1 to 2 seconds is 2, and the total number of data is 47. Since (1 + 2) /47=0.063..., The operation determination apparatus 101 sets the cumulative data ratio to 6%.

  Next, the operation determination apparatus 101 calculates the data number difference in the execution time frequency table 600. For example, the operation determination apparatus 101 calculates the data number difference by subtracting the data number of the record one level higher than the data number of the calculated data number difference record.

  In the example of FIG. 6, the operation determination apparatus 101 has a data number difference of 1 when the execution time is 0 to 1 second and the data number is 1 and when the execution time is 1 to 2 seconds. And The operation determination apparatus 101 sets the difference in the number of data to −1 because the number of data is 2 when the execution time is 1 to 2 seconds and the number of data is 1 when the execution time is 2 to 3 seconds.

  Next, the operation determination apparatus 101 sets a change in the difference in the execution time frequency table 600. For example, the operation determination apparatus 101 sets “increase” when the data number difference increases compared to the data number difference of the record immediately above, and “decrease” when the data number difference decreases. If the data number difference does not change, set blank.

  In the example of FIG. 6, the difference in the number of data when the execution time is 0 to 1 second is 1, the difference in the number of data is 1 when the execution time is 1 to 2 seconds, and the data number difference does not change. The apparatus 101 sets the change in the difference when the execution time is 1 to 2 seconds to be blank. Further, the difference in the number of data is 1 when the execution time is 1 to 2 seconds, and the difference in the number of data is −1 when the execution time is 2 to 3 seconds. The change in the difference in seconds is reduced.

  Next, the operation determination apparatus 101 detects a peak in which the change in the difference matches the pattern of “increase → decrease → increase” from the execution time frequency table 600 (step S804). For example, the operation determination apparatus 101 searches from a record with a short execution time, and searches for a record whose difference change is “increase”. Next, the operation determination apparatus 101 searches the record immediately below the record searched for “increase”, and searches for a record whose change in difference is “decrease”. Here, the next lower record is a record having an execution time longer by a certain time than the execution time of the record.

  Next, the operation determination apparatus 101 searches the record immediately below the record searched for “decrease”, and searches for a record whose change in difference is “increase”. The operation determination apparatus 101 sets one peak from the record that first searches for “increase” to the record that lastly searches for “increase”. The operation determining apparatus 101 repeats the same operation until the last record of the execution time frequency table 600 to detect a peak.

  In the example of FIG. 6, the operation determining apparatus 101 searches the record 600-1 as a record whose difference change is “increase”. Next, the operation determination apparatus 101 searches the record 600-3 as a record whose difference change is “decrease”. Next, the operation determination apparatus 101 searches the record 600-5 as a record whose difference change is “increase”. In this way, the operation determination apparatus 101 detects the records 600-1 to 600-5 as one peak. Similarly, the determination unit 702 detects records 600-7 to 600-10 and records 600-10 to 600-14 as peaks.

  Next, the operation determination apparatus 101 calculates the upper limit value of the execution time of the error operation from the execution time frequency table 600 (step S805). Here, the upper limit calculation processing procedure by the operation determination device 101 will be described later with reference to FIG.

  Next, the operation determination apparatus 101 extracts a peak corresponding to the calculated upper limit value of the error operation execution time (step S806). For example, the operation determination apparatus 101 extracts a peak including the upper limit value of the execution time of the error operation. For example, the operation determination apparatus 101 extracts the maximum peak that does not exceed the upper limit value of the execution time of the error operation.

  Next, the operation determination apparatus 101 extracts a record corresponding to the skirt in the extracted peak (step S807). For example, when the peak including the upper limit value of the execution time of the error operation is extracted, the operation determination device 101 extracts the first record in the extracted peak. For example, when the maximum peak that does not exceed the upper limit of the execution time of the error operation is extracted, after the change in the difference becomes “decrease” in the extracted peak, the first “increase” record is displayed. Extract.

  Finally, the operation determination apparatus 101 sets the execution time of the extracted record as a threshold value (step S808).

  In the example of FIG. 6, it is assumed that the upper limit value of the error operation execution time is calculated as “8” seconds in the upper limit value calculation processing procedure. In this case, the operation determination apparatus 101 detects the records 600-1 to 600-5 as peaks that do not exceed the upper limit “8” seconds of the operation execution time. The operation determining apparatus 101 detects the first “increase” record 600-5 after the change in the difference becomes “decrease” in the records 600-1 to 600-5. The operation determination apparatus 101 uses the execution time “5” seconds of the record 600-5 as a threshold value.

  The operation determining apparatus 101 determines whether or not the thresholds for all applications have been determined (step S809). When the threshold values for all applications have been determined (step S809: Yes), the processing of the operation determination apparatus 101 ends. When not determining the threshold values of all applications (step S809: No), the process of the operation determination apparatus 101 returns to step S804, and determines the threshold values of the applications for which the threshold values have not been determined.

  Thereby, a series of processing in this flowchart is completed. By executing this flowchart, the operation determination apparatus 101 calculates a threshold value for determining whether an application activation instruction is an intended operation or an unintended operation for each application.

(Example of upper limit calculation processing procedure)
FIG. 9 is a flowchart illustrating an example of the upper limit calculation process procedure of the operation determination apparatus 101. In FIG. 9, first, the operation determination apparatus 101 extracts the first record from the execution time frequency table 600 (step S901). The operation determining apparatus 101 determines whether or not the cumulative data number ratio of the extracted record> = the error ratio (step S902).

  If the cumulative data number ratio> = not the error ratio (step S902: No), the operation determination apparatus 101 extracts the next record of the execution time frequency table 600 (step S903). Thereafter, the process of the operation determination apparatus 101 returns to step S902.

  When the cumulative data number ratio> = the error ratio (step S902: Yes), the operation determining apparatus 101 sets the execution time of the extracted record as the upper limit value of the error operation execution time (step S904).

  In the example of FIG. 6, the error rate is 20%. In this case, since the operation determination device 101 has the cumulative data ratio of the record 600-8 being 28% and 20% or more, the execution time “8” seconds of the record 600-8 is set to the upper limit value of the error operation execution time. And

  Thereby, a series of processing in this flowchart is completed. By executing this flowchart, the operation determination apparatus 101 can calculate the upper limit value of the execution time of the error operation.

(Specific example of threshold determination)
FIG. 10 is an explanatory diagram illustrating a first specific example of threshold determination of the operation determination device 101. The operation determination apparatus 101 detects four peaks that match the pattern of “increase → decrease → increase” from the execution time frequency table 1000. Specifically, the four peaks are a record 1000-1 to a record 1000-3, a record 1000-3 to a record 1000-5, a record 1000-5 to a record 1000-9, and a record 1000-9 to a record 100-13. .

  Here, assuming that the error rate is 20%, the operation determination apparatus 101 calculates 6 seconds as the upper limit value of the error operation execution time from the execution time frequency table 1000. The operation determination apparatus 101 has records 1000-1 to 1000-3 and records 1000-5 that are the largest peaks among records 1000-1 to 1000-3 and records 1000-3 to 1000-5 that do not exceed 6 seconds. To extract.

  The operation determining apparatus 101 extracts the first “increase” record 1000-5 after the change in the difference becomes “decrease” among the extracted peaks, and sets the execution time of the record 1000-5 as 5 seconds as a threshold. And decide.

  In the example of FIG. 10, there are two peaks in the execution part due to the error operation. For example, one can be interpreted as a peak when an application is immediately stopped when the user notices that it is an unintended application when the application is started. Further, for example, the other one can be interpreted as a peak when the application is stopped after the application screen is displayed and the application is noticed as an unintended application when the application is started.

  FIG. 11 is an explanatory diagram illustrating a second specific example of threshold determination performed by the operation determination device 101. The operation determination apparatus 101 detects three peaks that match the pattern of “increase → decrease → increase” from the execution time frequency table 1100. Specifically, the three peaks are a record 1100-1 to a record 1100-4, a record 1100-4 to a record 1100-10, and a record 1100-10 to a record 1100-15.

  Here, assuming that the error rate is 20%, the operation determination apparatus 101 calculates 6 seconds as the upper limit value of the execution time of the error operation from the execution time frequency table 1100. The operation determination apparatus 101 extracts records 1100-1 to 1100-4 that are peaks not exceeding “6” seconds.

  The operation determining apparatus 101 extracts the first “increase” record 1100-4 after the change in the difference becomes “decrease” among the extracted peaks, and sets the execution time of the record 1100-4 as 4 seconds as a threshold. And decide.

  In the example of FIG. 11, there are a plurality of peaks, but there is one peak in the execution part due to an error operation. This is an example where it can be interpreted that there are multiple usages in normal use and multiple execution time peaks. For example, one can be interpreted as a peak when the user refers to information displayed when the user starts the application and then stops the application. Further, for example, the other one can be interpreted as a peak when, for example, the user refers to the information displayed when the user starts the application, and then stops the application after inputting information to the application.

  As described above, the operation determination apparatus 101 refers to the storage unit 701 that stores an application execution time history, and generates an execution time frequency distribution. Thereafter, the operation determining apparatus 101 determines the time at the point corresponding to the bottom of the peak including the predetermined time among the peaks of the frequency distribution of the generated execution time as the threshold value. Thereafter, the operation determination apparatus 101 determines whether the application start instruction is an intended operation or an unintended operation based on whether or not the execution time from the start of the application to the end is equal to or greater than a threshold value.

  Thereby, the operation determination apparatus 101 can make the operation included in the peak including the predetermined time the intended operation. For this reason, even if the predetermined time is increased in order to eliminate unintended operations, the operation determination device 101 determines that the intended operation is an unintended operation by setting the threshold of determination to the bottom of the peak. The number of correct operations does not decrease.

  As described above, the operation determination apparatus 101 can improve the performance of excluding the recording of an operation that is not intended by the user, so that the quality of the learning result of the operation of the user's application can be improved. As a result, the accuracy of the function of presenting an application that the user is likely to use can be improved, leading to an improvement in user convenience.

  Also, the predetermined time can satisfy that the number of data whose execution time is shorter than the predetermined time in the execution time frequency table 600 is equal to or less than a certain ratio with respect to the total number of data.

  Here, since there are individual differences in the reaction time of a person, the time for determining whether or not the application is activated by an unintended operation varies depending on the individual, and the execution time of the application by an unintended operation varies depending on the individual. For this reason, the operation determination apparatus 101 changes the threshold value for each user and determines that the intended operation is an unintended operation when the user inputs a certain rate at which the user makes an error in starting the application. Can be reduced.

  In addition, the operation determination apparatus 101 sets, as a threshold, a time at which a change in the difference of the execution time frequency table 600 increases from a decrease in a direction in which the execution time becomes longer in a maximum peak that does not exceed a predetermined time. Can be determined.

  Thereby, the operation determination apparatus 101 can reduce the removal of the intended operation at the same time as removing the data included in the peak not exceeding the predetermined time. Further, even when there is no peak including a predetermined time among the peaks of the frequency distribution of the created execution time, the operation determination apparatus 101 can determine the threshold value.

  Further, the operation determination device 101 refers to a storage unit that stores a history of execution times of a plurality of applications, generates a frequency distribution of execution times for each application of the plurality of applications, and determines a threshold value for each application. Can do.

  Here, the time for the user to determine whether the activated application is intended or not intended may vary depending on the application. For example, when a splash screen is displayed during activation after the application is activated, the execution time is short because the user immediately notices an error and terminates the application after completion of activation. On the other hand, for example, when the splash screen is not displayed during startup after the application is started, the execution time is long because the user does not notice an error until the application is started.

  As described above, the execution time of an application due to an unintended operation varies depending on the application. For this reason, by determining the threshold value for each application, the operation determination apparatus 101 can reduce the determination that the intended operation is an unintended operation.

  The operation determination method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. The operation determination program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. The operation determination program may be distributed via a network such as the Internet.

  The following additional notes are disclosed with respect to the embodiment described above.

(Supplementary Note 1) When determining whether an application activation instruction is an intended operation or an unintended operation based on whether the execution time from when the application is activated until it is terminated is equal to or greater than a threshold value,
Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time,
The time at the point corresponding to the bottom of the peak including a predetermined time among the generated peak of the frequency distribution of the execution time is determined as the threshold value.
An operation determination program for causing a computer to execute processing.

(Additional remark 2) The said predetermined time satisfy | fills that the number of data whose execution time is shorter than the said predetermined time is below a fixed ratio with respect to the number of all the data of the said log | history. The operation determination program according to 1.

(Supplementary Note 3) The determination process corresponds to a change in the number of data in the history from a decrease to an increase in a direction in which the execution time becomes longer from a peak located at the longest execution time within a range not exceeding the predetermined time. The operation determination program according to appendix 1 or 2, wherein a point time is determined as the threshold value.

(Appendix 4) The storage unit stores a history of execution times of a plurality of applications,
The generating process generates a frequency distribution of execution times for each of the plurality of applications,
The operation determination program according to any one of appendices 1 to 3, wherein the determining process determines the threshold value for each application.

(Supplementary Note 5) Determining whether an application activation instruction is an intended operation or an unintended operation from whether the execution time from the activation of the application to the termination is equal to or greater than a threshold corresponding to the application The operation determination program according to any one of Supplementary notes 1 to 4, which is characterized by the following.

(Supplementary Note 6) When determining whether the application start instruction is an intended operation or an unintended operation from whether the execution time from the start of the application to the end is equal to or greater than a threshold,
Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time,
The time at the point corresponding to the bottom of the peak including a predetermined time among the generated peak of the frequency distribution of the execution time is determined as the threshold value.
An operation determination program for causing a computer to execute processing is recorded. The computer-readable recording medium.

(Supplementary note 7) When determining whether the application start instruction is an intended operation or an unintended operation from whether the execution time from the start of the application to the end is equal to or greater than a threshold,
Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time,
The time at the point corresponding to the bottom of the peak including a predetermined time among the generated peak of the frequency distribution of the execution time is determined as the threshold value.
An operation determination method, wherein a computer executes a process.

(Supplementary Note 8) When determining whether the application start instruction is an intended operation or an unintended operation based on whether the execution time from the start of the application to the end is equal to or greater than a threshold,
Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time,
The time at the point corresponding to the bottom of the peak including a predetermined time among the generated peak of the frequency distribution of the execution time is determined as the threshold value.
An operation determination apparatus having a control unit.

101 Operation Determination Device 701 Storage Unit 702 Determination Unit 703 Determination Unit

Claims (6)

When determining whether the application start instruction is an intended operation or an unintended operation from whether the execution time from the start of the application to the end is equal to or greater than a threshold,
Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time,
The time at the point corresponding to the bottom of the peak including a predetermined time among the generated peak of the frequency distribution of the execution time is determined as the threshold value.
An operation determination program for causing a computer to execute processing.   2. The predetermined time satisfies the condition that the number of data whose execution time is shorter than the predetermined time is equal to or less than a predetermined ratio with respect to the total number of data in the history. Operation judgment program.   In the process of determining, the time at which the change in the number of data in the history corresponds to the increase from the decrease in the direction in which the execution time becomes longer from the peak located at the longest execution time within the range not exceeding the predetermined time. The operation determination program according to claim 1, wherein the threshold value is determined as the threshold value. The storage unit stores a history of execution times of a plurality of applications,
The generating process generates a frequency distribution of execution times for each of the plurality of applications,
The operation determination program according to claim 1, wherein the determining process determines the threshold value for each application. When determining whether the application start instruction is an intended operation or an unintended operation from whether the execution time from the start of the application to the end is equal to or greater than a threshold,
Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time,
The time at the point corresponding to the bottom of the peak including a predetermined time among the generated peak of the frequency distribution of the execution time is determined as the threshold value.
An operation determination method, wherein a computer executes a process. When determining whether the application start instruction is an intended operation or an unintended operation from whether the execution time from the start of the application to the end is equal to or greater than a threshold,
Referring to the storage unit that stores the history of the execution time of the application, generate a frequency distribution of the execution time,
The time at the point corresponding to the bottom of the peak including a predetermined time among the generated peak of the frequency distribution of the execution time is determined as the threshold value.
An operation determination apparatus having a control unit.

Images