JP7222688B2 - Information processing terminal, information processing method, and information processing program - Google Patents

Description

The present disclosure relates to technology for assisting improvement of work efficiency.

In recent years, RPA (Robotic Process Automation) technology has attracted attention. RPA technology is a technology for automating repetitive routine tasks. Work efficiency is improved by automating routine tasks.

Regarding RPA technology, Non-Patent Document 1 discloses a business automation support tool. The business automation support tool aggregates the user's operation log, creates a graph for each business content or each application in order of work time, and reports the graph.

Tenda Co., Ltd., "D-Analyzer", [online], [searched on December 10, 2018], Internet (https://www.d-analyzer.jp/feature/)

In order to improve work efficiency, it is important to grasp a series of tasks that are performed repeatedly. Therefore, it is not enough to pay attention only to the work time of specific business content like the business automation support tool shown in Non-Patent Document 1. Therefore, it is desired to help improve work efficiency in a non-conventional way.

In one example of the present disclosure, an information processing terminal includes a communication unit that communicates with a plurality of business terminals. The communication unit receives an operation log of the business terminal from each of the plurality of business terminals. In the operation log, each piece of operation information for the business terminal of the transmission source is specified in chronological order. The information processing terminal further includes a work identification unit for identifying work content corresponding to each piece of operation information from time-series operation information defined in each operation log received, and A repetition identification unit for identifying a series of repetitive work that is being performed repeatedly from each work content identified for each piece of time-series operation information; and an output unit for

In one example of the present disclosure, each piece of operation information defined in each received operation log is further associated with the work time spent on the operation. The output unit calculates an average work time spent on the work content for each work content that constitutes the repetitive work based on each work time specified in each of the received operation logs, and calculates It also outputs each average work time.

In one example of the present disclosure, the output unit calculates the total work time spent on the repeated work based on each work time defined in each of the received operation logs, and further outputs the total work time. do.

In an example of the present disclosure, when a plurality of patterns of repetitive work are identified by the repetition identification unit, the output unit calculates the total work time for each of the plurality of patterns of repetitive work, and calculates the total work time for each of the plurality of patterns of repetitive work. and a graph representing in a comparable manner the total work time for each of the .

In an example of the present disclosure, each piece of operation information defined in each received operation log is further associated with user identification information indicating the user who performed the operation. The output unit calculates the total number of users who have performed the repeated work based on the user identification information defined in each of the received operation logs, and further outputs the total number.

In one example of the present disclosure, the output unit further outputs a graph representing the work amount of the repetitive work in a comparable manner for each user.

In an example of the present disclosure, the output unit identifies the number of times the repeated work is included in each work content identified for each piece of time-series operation information by the work identification unit, and sets the identified number to the above It is also output as the number of iterations performed.

In an example of the present disclosure, each piece of operation information defined in each received operation log is further associated with an operation time at which the operation was performed. The output unit calculates, for each predetermined time segment, the frequency of the repeated work performed based on each operation time specified in each of the received operation logs, and for each predetermined time segment , a graph showing the frequency is further output.

In one example of the present disclosure, an information processing method includes receiving, from each of a plurality of business terminals, an operation log of the business terminal. In the operation log, each piece of operation information for the business terminal of the transmission source is specified in chronological order. The information processing method further comprises a step of identifying work content corresponding to each piece of operation information from time-series pieces of operation information defined in each operation log received; A step of specifying a series of repetitive work that is performed repeatedly from each work content specified for operation information, and a step of arranging and outputting each work content constituting the repetitive work in chronological order.

In one example of the present disclosure, an information processing program causes a computer to execute a step of receiving an operation log of a business terminal from each of a plurality of business terminals. In the operation log, each piece of operation information for the business terminal of the transmission source is specified in chronological order. The information processing program further instructs the computer to identify work content corresponding to each piece of operation information from time-series pieces of operation information defined in each operation log received; a step of identifying a series of repetitive work that is being performed repeatedly from each work content identified for each piece of time-series operation information; and a step of arranging and outputting each work content constituting the repetitive work in chronological order to run.

It is a figure which shows an example of the apparatus configuration of an information processing system. FIG. 4 is a diagram showing an example of a functional configuration for identifying repeated work from an operation log; FIG. 3 is a schematic diagram showing an example of a hardware configuration of a work terminal; FIG. It is a schematic diagram which shows an example of the hardware constitutions of an information processing terminal. It is a figure which shows an example of the functional structure of a work specification part. FIG. 10 is a diagram illustrating an example of a functional configuration of a repetition specifying unit; It is a figure which shows the process which produces|generates the data set for learning. FIG. 4 is a diagram schematically showing the process of learning processing by a learning unit; FIG. 4 is a diagram schematically showing the process of analysis processing by an analysis unit; It is a figure which shows an example of the functional structure of an output part. FIG. 4 is a diagram showing a process of calculating an average work time of each work that constitutes repetitive work; FIG. 5 is a diagram showing a process of calculating a total work time spent on repeated work; FIG. 10 is a diagram showing a process of calculating the number of users who have performed repeated tasks; It is a figure which shows the process of calculating the frequency|count of repeating work. FIG. 10 is a diagram showing a work report according to specific example 1; FIG. 11 is a diagram showing a work report according to specific example 2; FIG. 12 is a diagram showing a work report according to specific example 3; FIG. 13 is a diagram showing a work report according to specific example 4; 4 is a flowchart showing a part of processing executed by a work terminal; 4 is a flowchart showing a part of processing executed by an information processing terminal; It is a figure which shows roughly the function of the learning data production|generation part according to a modification. FIG. 10 is a diagram showing a working terminal according to a modified example;

Hereinafter, each embodiment according to the present invention will be described with reference to the drawings. In the following description, identical parts and components are given identical reference numerals. Their names and functions are also the same. Therefore, detailed description of these will not be repeated.

<A. Information processing system 500>
An information processing system 500 will be described with reference to FIG. FIG. 1 is a diagram showing an example of the device configuration of an information processing system 500. As shown in FIG.

As shown in FIG. 1 , the information processing system 500 includes multiple work terminals 100 and one or more information processing terminals 200 .

Work terminal 100 is, for example, a notebook or desktop PC (Personal Computer), tablet terminal, smart phone, or other information processing terminal having a communication function. The number of work terminals 100 constituting the information processing system 500 is arbitrary, and is, for example, several to several thousand. For simplicity of illustration, only three workstations 100 are shown in FIG. Each of the work terminals 100 may be placed in the same building or may be placed in different buildings.

The work terminal 100 is a terminal for a user to work. An operation monitoring program 122 (see FIG. 2), which will be described later, is installed in the work terminal 100 to monitor user operations. The operation monitoring program 122 writes user operations on the work terminal 100 into the operation log 130 and periodically transmits the operation log 130 to the information processing terminal 200 .

The information processing terminal 200 is, for example, a notebook or desktop PC, tablet terminal, smart phone, or other information processing terminal having a communication function. Although one information processing terminal 200 is shown in FIG. 1, the number of information processing terminals 200 constituting the information processing system 500 is arbitrary.

The information processing terminal 200 analyzes the operation log 130 collected from each of the work terminals 100 and reports various indicators for business improvement.

<B. Overview of Information Processing Terminal 200>
Next, an outline of the information processing terminal 200 described above will be described with reference to FIG. FIG. 2 is a diagram showing an example of a functional configuration for identifying repeated work from the operation log 130. As shown in FIG.

As shown in FIG. 2 , information processing terminal 200 includes communication section 252 , work identification section 254 , repetition identification section 256 , and output section 260 .

The communication unit 252 is, for example, a communication driver that provides communication functions with the work terminal 100 . The communication unit 252 communicates with the work terminal 100 via, for example, the Internet, a wide area network (WAN), a local area network (LAN), an Ethernet (registered trademark) network, a public or private network. to achieve communication with

The communication unit 252 receives the operation log 130 of the work terminal 100 from each of the plurality of work terminals 100 . In the operation log 130, each piece of information (hereinafter also referred to as “operation information”) regarding an operation performed on the work terminal 100 that is the transmission source is defined in chronological order. The “operation information” indicates, for example, data on which the user has operated within the work terminal 100 . The "operation information" in this specification includes the type of window to be operated, the type of data to be operated, the type of application to be operated, the UI (User Interface) component to be operated, and the screen to be operated (for example, a captured screen), coordinate information indicated by an input device such as a mouse, key input results on a keyboard, or a combination thereof. Communication unit 252 outputs each received operation log 130 to work identification unit 254 .

The work identification unit 254 identifies the work content corresponding to each piece of operation information from each piece of time-series operation information defined in each piece of the operation log 130 . The work content corresponding to each piece of operation information is specified based on the work definition 232, for example. The work definition 232 associates work content with the type of operation information. As an example, the work identification unit 254 refers to the work definition 232 and identifies work “A” as a corresponding work for the operation target “A” defined in the operation log 130 . Further, the work identifying unit 254 identifies work "B" as a corresponding work for the operation target "C".

Note that the corresponding work content does not necessarily have to be specified based on the work definition 232 . For example, the work identifying unit 254 may delete unnecessary keywords from each piece of time-series operation information defined in each of the operation logs 130 and identify the edited operation information as the corresponding work content. The work identification result by the work identification unit 254 is output to the repeat identification unit 256 .

The repetitive identification unit 256 identifies a series of repetitive tasks (that is, repetitive tasks) from each work content identified for each piece of time-series operation information by the work identification unit 254 . In other words, the repetitive identifying unit 256 identifies a group of tasks with similar order of work contents as repetitive tasks from among the results of task identification by the task identifying unit 254 . In the example of FIG. 2, since a series of work "Work C→Work B→Work D" is repeatedly performed, the repetition identifying unit 256 identifies "Work C→Work B→Work D" as a repeated work. . The repetition identification unit 256 outputs the identified repetition work to the output unit 260 .

In the example of FIG. 2, an example in which repetitive work of one type of work pattern is specified is shown, but repetitive work of a plurality of types of work patterns may be specified.

The output unit 260 arranges and outputs the work contents constituting the identified repetitive work in chronological order. The repetitive work is output by any output means. As an example, the output unit 260 outputs repetitive work as a work report 270 . In the example of FIG. 2, in the work report 270, each work content that constitutes repeated work is connected by arrows. Details of the work report 270 will be described later.

By checking the work report 270, the user can grasp a series of work that is being performed repeatedly. By presenting such repetitive work, the user can easily discover work content that should be improved. This advantage becomes more pronounced as the business becomes more complex. The user can greatly improve work efficiency by automating the presented repetitive work. In this way, the information processing terminal 200 can help improve the work efficiency of the work terminal 100 .

<C. Hardware Configuration>
The hardware configurations of work terminal 100 and information processing terminal 200 shown in FIG. 1 will be described in order with reference to FIGS. 3 and 4. FIG.

(C1. Hardware configuration of work terminal 100)
First, the hardware configuration of the work terminal 100 will be described with reference to FIG. FIG. 3 is a schematic diagram showing an example of the hardware configuration of the work terminal 100. As shown in FIG.

The work terminal 100 includes a control device 101 , a ROM (Read Only Memory) 102 , a RAM (Random Access Memory) 103 , a communication interface 104 , a display interface 105 , an input interface 107 and a storage device 120 . These components are connected to bus 110 .

The control device 101 is composed of, for example, at least one integrated circuit. Integrated circuits include, for example, at least one CPU (Central Processing Unit), at least one GPU (Graphics Processing Unit), at least one ASIC (Application Specific Integrated Circuit), at least one FPGA (Field Programmable Gate Array), or It can be configured by a combination of

The control device 101 controls the operation of the work terminal 100 by executing various programs such as the operation monitoring program 122 and work program 132 of the work terminal 100 and the operating system. The operation monitoring program 122 is a program for monitoring operations on the work terminal 100. The operation monitoring program 122 has a function of writing information related to user operations into the operation log 130, and periodically writing the operation log 130 to the information processing terminal 200 (see FIG. 1). including the ability to send to The work program 132 is an application for supporting the work of the user. As an example, the work program 132 includes various mailers, various browsers, Word, Excel (registered trademark), and other applications that support business.

The control device 101 reads the operation monitoring program 122 and the work program 132 from the storage device 120 or the ROM 102 to the RAM 103 based on the reception of the instruction to execute the operation monitoring program 122 . The RAM 103 functions as a working memory and temporarily stores various data necessary for executing the operation monitoring program 122 and the working program 132 .

A LAN (Local Area Network), an antenna, and the like are connected to the communication interface 104 . The work terminal 100 exchanges data with external devices via the communication interface 104 . The external device includes, for example, the information processing terminal 200 described above, a server, and the like. The work terminal 100 may be configured so that the operation monitoring program 122 and work program 132 can be downloaded from an external device such as a server.

A display 106 is connected to the display interface 105 . Display interface 105 sends an image signal for displaying an image to display 106 in accordance with a command from control device 101 or the like. The display 106 is, for example, a liquid crystal display, an organic EL (Electro Luminescence) display, or other display device. Note that the display 106 may be configured integrally with the work terminal 100 or may be configured separately from the work terminal 100 .

An input device 108 is connected to the input interface 107 . Input device 108 is, for example, a mouse, keyboard, touch panel, or other device capable of receiving user operations. Note that the input device 108 may be configured integrally with the work terminal 100 or may be configured separately from the work terminal 100 .

The storage device 120 is, for example, a storage medium such as a hard disk or flash memory. The storage device 120 stores an operation monitoring program 122, an operation log 130, a work program 132, and the like. Note that the storage location of the operation monitoring program 122, the operation log 130, and the work program 132 is not limited to the storage device 120, but can be a storage area (for example, cache memory) of the control device 101, ROM 102, RAM 103, external devices ( For example, it may be stored in a server) or the like.

The operation monitoring program 122 may be provided not as a standalone program but as part of any program. In this case, the communication monitoring process by the operation monitoring program 122 is implemented in cooperation with any program. Even a program that does not include such a part of modules does not deviate from the gist of the operation monitoring program 122 according to the present embodiment. Furthermore, part or all of the functions provided by the operation monitoring program 122 may be realized by dedicated hardware. Further, the work terminal 100 may be configured as a so-called cloud service in which at least one server executes part of the processing of the operation monitoring program 122 .

(C2. Hardware configuration of information processing terminal 200)
Next, a hardware configuration of the information processing terminal 200 will be described with reference to FIG. FIG. 4 is a schematic diagram showing an example of the hardware configuration of the information processing terminal 200. As shown in FIG.

Information processing terminal 200 includes control device 201 , ROM 202 , RAM 203 , communication interface 204 , display interface 205 , input interface 207 , and storage device 220 . These components are connected to bus 210 .

The control device 201 is, for example, composed of at least one integrated circuit. An integrated circuit may be comprised of, for example, at least one CPU, at least one GPU, at least one ASIC, at least one FPGA, or combinations thereof.

The control device 201 controls the operation of the information processing terminal 200 by executing various programs such as an information processing program 222 and an operating system. The control device 201 reads the information processing program 222 from the storage device 220 or the ROM 202 to the RAM 203 based on the reception of the execution command of the information processing program 222 . The RAM 203 functions as a working memory and temporarily stores various data necessary for executing the information processing program 222 .

A LAN, an antenna, and the like are connected to the communication interface 204 . Information processing terminal 200 exchanges data with external devices via communication interface 204 . The external device includes, for example, the work terminal 100 described above, a server, and the like. The information processing terminal 200 may be configured so that the information processing program 222 can be downloaded from the external device.

A display 206 is connected to the display interface 205 . Display interface 205 sends an image signal for displaying an image to display 206 in accordance with a command from control device 201 or the like. The display 206 is, for example, a liquid crystal display, organic EL display, or other display device. Note that the display 206 may be configured integrally with the information processing terminal 200 or may be configured separately from the information processing terminal 200 .

An input device 208 is connected to the input interface 207 . Input device 208 is, for example, a mouse, keyboard, touch panel, or other device capable of receiving user operations. Note that the input device 208 may be configured integrally with the information processing terminal 200 or may be configured separately from the information processing terminal 200 .

The storage device 220 is, for example, a storage medium such as a hard disk or flash memory. The storage device 220 stores an information processing program 222, the work definition 232 described above, the work report 270 described above, a work estimation model M described later, and the like. These storage locations are not limited to storage device 220, and may be stored in a storage area of control device 201 (for example, cache memory), ROM 202, RAM 203, external equipment (for example, server), or the like.

The information processing program 222 may be provided not as a standalone program but as part of an arbitrary program. In this case, the operation log analysis processing by the information processing program 222 is realized in cooperation with an arbitrary program. Even a program that does not include such a part of modules does not deviate from the gist of information processing program 222 according to the present embodiment. Furthermore, part or all of the functions provided by the information processing program 222 may be realized by dedicated hardware. Furthermore, the information processing terminal 200 may be configured as a so-called cloud service in which at least one server executes part of the processing of the information processing program 222 .

<D. Function of Work Identification Unit 254>
Next, with reference to FIG. 5, the details of the function of the work identification unit 254 shown in FIG. 2 will be described. FIG. 5 is a diagram showing an example of the functional configuration of the work identification unit 254. As shown in FIG.

The work identification unit 254 receives the input of the operation log 130 and the work definition 232 and outputs a work identification result D1.

The operation log 130 contains various information related to operations on the work terminal 100 . Various types of information included in the operation log 130 are arranged in the order in which operations were performed on the work terminal 100 . The operation log 130 includes, for example, terminal identification information indicating the work terminal 100 to be operated, user identification information indicating the user who performed the operation, operation time indicating the timing at which the operation was performed, and time spent on the operation. It includes work time indicating time, application identification information indicating an application to be operated, and operation information for the source work terminal 100 .

In the example of FIG. 5, a personal computer name is shown as an example of the terminal identification information, but the terminal identification information may be indicated by IP (Internet Protocol) address, MAC (Media Access Control) address, or the like. Also, in the example of FIG. 5, a login name is shown as an example of user identification information, but user identification information may be shown by a user ID (Identification), an IP address, a MAC address, or the like. Also, in the example of FIG. 5, an application name is shown as an example of the application identification information, but the application identification information may be shown by a file path or the like indicating the location of the EXE file. In the example of FIG. 5, the title of the window to be operated is shown as an example of the operation information, but the operation information includes the type of data to be operated (for example, file name and file path), It may be defined by UI components (for example, button names, etc.). Alternatively, the operation information may be defined by a screen to be operated (for example, a captured screen), coordinate information indicated by an input device such as a mouse, key input results for input using a keyboard, or the like.

The work definition 232 associates the work content (for example, the title of the window to be operated) with the type of operation information. In the example of FIG. 5, the title of the window to be operated is shown as an example of the operation information defined in the work definition 232, but the operation information is the type of data to be operated (for example, file name and file path). ), a UI component to be operated (for example, a button name), or the like.

Based on the work definition 232 , the work identification unit 254 identifies the corresponding work content for each piece of time-series operation information defined in the operation log 130 . More specifically, when the operation information specified in the operation log 130 is referred to as first operation information, and the operation information specified in the work definition 232 is referred to as second operation information, the work identification unit 254 identifies a certain first operation information. Operation information is used as an identification target, and second operation information that matches the first operation information to be identified is specified from among the second operation information. Then, the work identifying unit 254 sets the work content associated with the matching second operation information as the work content corresponding to the first operation information to be identified. The work identification unit 254 executes such work content identification processing for all pieces of first operation information defined in the operation log 130 .

It should be noted that the concept of "matching" here can include not only complete matching of the first operation information and the second operation information, but also approximate matching of the first operation information and the second operation information. For example, a case where one of the first operation information and the second operation information partially matches the other may be included in the concept of "match". As an example, the work identifying unit 254 identifies second operation information that matches a part of the first operation information to be identified. Then, the work identification unit 254 sets the work content associated with the identified second operation information as the work content corresponding to the first operation information to be identified.

As a result of the work identification processing by the work identification unit 254, a work identification result D1 is output. As described above, the operation information defined in the operation log 130 is arranged in chronological order (that is, the order of operations), so the work content group defined in the work identification result D1 is also in chronological order (that is, , order of operation). The work identification result D<b>1 is output to the repeat identification unit 256 .

<E. Function of Repeat Identification Unit 256>
Next, with reference to FIGS. 6 to 9, the details of the function of the repetition identification unit 256 shown in FIG. 2 will be described. FIG. 6 is a diagram showing an example of the functional configuration of the repetition specifying unit 256. As shown in FIG.

As shown in FIG. 6, the repetition identification unit 256 includes a learning data generation unit 257, a learning unit 258, and an analysis unit 259 as functional modules. These functional modules will be described in turn below.

(E1. Learning data generator 257)
First, the function of the learning data generator 257 will be described with reference to FIG. FIG. 7 is a diagram showing the process of generating the learning data set D2.

As shown in FIG. 7, the learning data generation unit 257 receives an input of the work identification result D1 and generates a learning data set D2 used by the learning unit 258, which will be described later. More specifically, first, the learning data generation unit 257 divides the time-series work content group included in the work identification result D1 into predetermined units (for example, 10 units). The number of units is arbitrary. In the example of FIG. 7, three work contents specified in the work identification result D1 are blocked. Next, the learning data generation unit 257 tags the work content group included in each block as correct information with the work content group included in the next block following the block. In the example of FIG. 7, for example, the work content group included in the block of work pattern 1, which is included in the block of work pattern 1, is tagged with the work content group included in the block of work pattern 4 following work pattern 1 as correct information. Such tagging is performed for all blocks and they are generated as the training data set D2.

The learning data set D<b>2 generated by the learning data generating section 257 is output to the learning section 258 .

(E2. Learning unit 258)
Next, the function of the learning section 258 shown in FIG. 6 will be described with reference to FIG. FIG. 8 is a diagram schematically showing the process of learning processing by the learning section 258. As shown in FIG.

The learning unit 258 generates the work estimation model M by a predetermined machine learning process using the learning data set D2 generated by the learning data generation unit 257. FIG. The task estimation model M receives a series of task group inputs and outputs task groups following those task groups as estimation results. The machine learning process performed by the learning unit 258 may be deep learning, RNN (Recurrent Neural Network), support vector machine, or other learning. It may be a method. In the following, a method for generating a work estimation model in RNN will be described as an example.

FIG. 8 shows the work estimation model M1 in the initial state and the work estimation model M2 after the first parameter update. Hereinafter, the work estimation models M1 and M2 are also referred to as a work estimation model M when they are not distinguished from each other. The task estimation model M is composed of an input layer X, an intermediate layer H, and an output layer Y.

The input layer X receives input of work patterns (that is, input feature values) included in the learning data set D2. The input layer X outputs to the intermediate layer H each work content that constitutes the input work pattern.

The intermediate layer H is composed of a plurality of layers. The number of layers of the intermediate layer H is arbitrary. Each layer is composed of multiple units. Each unit constituting each layer of the intermediate layer H is connected to each unit of the previous layer and each unit of the next layer. Each unit in each layer receives each output value from each unit in the previous layer, multiplies each output value by a weight, integrates the multiplication results, and adds a predetermined bias to the integration result ( or subtraction), the addition result (or subtraction result) is input to a predetermined function (for example, sigmonite function), and the output value of the function is output to each unit in the next layer.

The output layer Y is composed of a plurality of units. Each unit forming the output layer Y is connected to each unit in the last layer of the intermediate layer H. FIG. Each unit of the output layer Y receives the output value from each unit of the final layer of the intermediate layer H, multiplies each output value by a weight, integrates the multiplication results, and biases the integration result with a predetermined bias. is added (or subtracted), the addition result (or subtraction result) is input to a predetermined function (for example, a sigmonite function), and the output result of the function is output as an output value. The output value indicates the type of work content. The work estimation model M estimates each work content output from the output layer Y as a subsequent work pattern following the input work pattern.

As a more specific example, the learning unit 258 inputs the first work pattern "1" defined in the learning data set D2 to the work estimation model M, and the resulting estimated work pattern "α". and work pattern "4" which is associated with work pattern "1" as correct information. Then, learning unit 258 updates various parameters (for example, weight and bias) included in work estimation model M so that estimated work pattern “α” approaches work pattern “4” as correct information.

Next, the learning unit 258 inputs to the work estimation model M the second work pattern “2” defined in the learning data set D2. In the second and subsequent learnings, the learning unit 258 further inputs to the input layer X part of the output of the intermediate layer H in the previous learning process. Thus, in the RNN, the output of one layer is not only used as the input of the next layer, but also input to the input layer X in the next learning process. As a result, the output result of the work estimation model M is affected not only by the second work pattern "2" but also by the first work pattern "1". As a result, temporal information is passed.

The learning unit 258 compares the estimated work pattern “β” obtained from the learning data set D2 with the work pattern “5” associated with the work pattern “2” as correct information. Then, learning unit 258 updates various parameters (for example, weight and bias) included in work estimation model M so that estimated work pattern “β” approaches work pattern “2” as correct information.

The learning unit 258 repeatedly performs update processing of the parameters included in the work estimation model M for each piece of learning data in the learning data set D2. As a result, the work estimation accuracy of the work estimation model M is improved, and when the work estimation model M receives the input of a series of work groups, the work estimation model M outputs work groups following the work group.

Note that the learning unit 258 does not necessarily need to use all the learning data included in the learning data set D2 in order to generate the task estimation model M. The data may be used to generate a task estimation model M.

(E3. Analysis unit 259)
Next, with reference to FIG. 9, functions of the analysis unit 259 shown in FIG. 6 will be described. FIG. 9 is a diagram schematically showing the process of analysis processing by the analysis unit 259. As shown in FIG.

The analysis unit 259 inputs the learning data set D2 again to the work estimation model M that has been learned by the learning unit 258, and identifies work patterns that are repetitive work.

More specifically, the analysis unit 259 inputs the first work pattern “1” defined in the learning data set D2 to the learned work estimation model M, and the resulting estimated work pattern “ γ” is compared with work pattern “4” associated with work pattern “1” as correct information to determine whether or not the estimated work pattern “γ” is similar to work pattern “4”. to decide. As an example, the analysis unit 259 calculates the correlation value by calculating the inner product of the estimated work pattern “γ” and the work pattern “4”. Then, when the correlation value exceeds the predetermined value, the analysis unit 259 regards work pattern “1” (or work pattern “4”) as repetitive work.

Similarly, the analysis unit 259 inputs the second work pattern “2” defined in the learning data set D2 to the learned work estimation model M, and the resulting estimated work pattern “δ” , and the work pattern "5" associated as correct information are compared with the work pattern "2" to determine whether or not the estimated work pattern "δ" is similar to the work pattern "5". As an example, the analysis unit 259 calculates the correlation value by calculating the inner product of the estimated work pattern “δ” and the work pattern “5”. Then, when the correlation value exceeds the predetermined value, the analysis unit 259 regards work pattern “2” (or work pattern “5”) as repetitive work.

The analysis unit 259 performs analysis processing of such repetitive work for all the learning data in the learning data set D2. As a result, repetitive tasks 269 are identified. The work pattern of the repetitive work identified at this time is not limited to one type, and a plurality of types of work patterns may be identified.

Note that when the work estimation model M is generated using part of the learning data set D2, the analysis unit 259 may input the rest of the learning data set D2 to the work estimation model M. . As a result, it is determined whether or not the rest of the learning data set D2 includes repeated work. After that, the analysis unit 259 changes the combination of the learning data set used for generating the task estimation model M and the remaining learning data sets used for analysis, thereby obtaining all the learning data of the learning data set D2. Repeated work analysis processing is performed for . This suppresses over-learning.

<F. Function of Output Unit 260>
Next, with reference to FIGS. 10 to 14, the details of the functions of the output section 260 shown in FIG. 2 will be described. FIG. 10 is a diagram showing an example of the functional configuration of the output unit 260. As shown in FIG.

As shown in FIG. 10, the output unit 260 includes an average time calculation unit 261, a total time calculation unit 262, a user number calculation unit 263, and a repetition number calculation unit 264 as functional modules. These functional modules will be described in turn below.

(F1. Average time calculation unit 261)
First, with reference to FIG. 11, the function of the average time calculator 261 will be described. FIG. 11 is a diagram showing the process of calculating the average work time of each work that constitutes repetitive work.

The average time calculation unit 261 determines the work pattern indicated by the repetitive work 269 identified by the repetitive identification unit 256 from among the time-series work contents specified in the above-described work identification result D1 (or the operation log 130). Extract. As an example, it is assumed that the identified repetitive work 269 includes the repetitive pattern “work E→work A→work G”. In this case, the average time calculation unit 261 extracts the repetition pattern “work E→work A→work G” from the time-series work contents defined in the work identification result D1.

As described above, in the work identification result D1, the work time spent on each work content is associated with each work content. Based on each of the work times, the average time calculation unit 261 calculates the average work time spent on the work content for each work content that constitutes the repetitive work. In the example of the repetitive work "Work E→Work A→Work G" shown in FIG. is calculated as the average work time of work "E". Similarly, the average time calculation unit 261 calculates the average time "TB" of the work times "ΔT32", "ΔT42", and "ΔT52" associated with work "A" as the average work time of work "A". do. Similarly, the average time calculation unit 261 calculates the average time "TC" of the work times "ΔT33", "ΔT43", and "ΔT53" associated with the work "G" as the average work time of the work "G". do.

The average time calculation unit 261 calculates such average work time calculation processing for each of the repetitive tasks 269 identified by the repetitive identification unit 256 described above. These calculated average work times are output to the report generator 265 .

(F2. Total time calculation unit 262)
Next, with reference to FIG. 12, functions of the total time calculation unit 262 shown in FIG. 10 will be described. FIG. 12 is a diagram showing the process of calculating the total work time spent on repeated work.

The total time calculation unit 262 extracts the work pattern indicated by the repetitive work 269 identified by the repetitive identification unit 256 from the time-series work content defined in the above-described work identification result D1 (see FIG. 5). do. As an example, it is assumed that the identified repetitive work 269 includes the repetitive pattern “work E→work A→work G”. In this case, the total time calculation unit 262 extracts the repetition pattern “Work E→Work A→Work G” from the time-series work contents specified in the work identification result D1.

As described above, in the work identification result D1, the work time spent on each work content is associated with each work content. Based on each of the work times, the total time calculation unit 262 accumulates the work time associated with the work content that constitutes the repetitive work, and uses the result of the accumulation as the total work time spent on the repetitive work. calculate. In the example of the repetitive work “work E→work A→work G” shown in FIG. The total time ``Tsum'' of ``ΔT31'' to ``ΔT33'', ``ΔT41'' to ``ΔT43'', and ``ΔT51'' to ``ΔT53'' is repeated. Calculate as

The total time calculation unit 262 calculates such a total work time calculation process for each of the repetitive tasks 269 identified by the repetitive identification unit 256 described above. Further, the total time calculation unit 262 calculates the total total work time of the repetitive work 269 without distinguishing the work patterns of the repetitive work. These calculated total work times are output to the report generator 265 .

(F3. Number of Users Calculation Unit 263)
Next, with reference to FIG. 13, functions of the number-of-users calculation unit 263 shown in FIG. 10 will be described. FIG. 13 is a diagram showing the process of calculating the number of users who have performed repeated tasks.

The number-of-users calculation unit 263 determines a work pattern indicated by the repetitive work 269 identified by the repetitive identification unit 256 from among the time-series work contents defined in the above-described work identification result D1 (or the operation log 130). Extract. As an example, it is assumed that the identified repetitive work 269 includes the repetitive pattern “work E→work A→work G”. In this case, the number-of-users calculation unit 263 extracts the repetition pattern “work E→work A→work G” from the time-series work contents specified in the work identification result D1.

As described above, the work identification result D1 is associated with the user identification information (for example, login name) for identifying the user who performed the work for each work content. The number-of-users calculation unit 263 calculates the total number of users who have repeatedly performed the work based on the user identification information. In the example of FIG. 13, two login names "User1" and one login name "User2" are associated with the repetitive work "work E→work A→work G". The number-of-users calculation unit 263 counts the number of users by eliminating duplication of the login name “User1”. As a result, the total number of users who performed the repetitive work "Work E→Work A→Work G" is calculated as two. In other words, the number-of-users calculation unit 263 calculates the number of types of user identification information associated with repetitive work as the total number of users.

The number-of-users calculation unit 263 calculates such calculation processing of the total number of users for each of the repetitive tasks 269 identified by the repetitive identification unit 256 described above. Further, the user number calculation unit 263 calculates the total number of users who have performed the repetitive work 269 without distinguishing between work patterns of the repetitive work. These calculated user totals are output to the report generation unit 265 .

(F4. Repetition number calculation unit 264)
Next, with reference to FIG. 14, the function of the repetition count calculator 264 shown in FIG. 10 will be described. FIG. 14 is a diagram showing the process of calculating the number of repeated tasks.

The repetition count calculation unit 264 identifies the number of times the repeated work 269 is included in the time-series work content defined in the work identification result D1 (see FIG. 5), and repeats the identified number. It is calculated as the number of times the work has been performed (hereinafter also referred to as “repeated count”). As an example, it is assumed that the repeating pattern “work E→work A→work G” is included in the identified repetitive work 269 . In this case, the repeat count calculation unit 264 extracts the repeat work “work E→work A→work G” from the time-series work contents specified in the work identification result D1. As a result, it is assumed that three repetitive tasks "Work E→Work A→Work G" are extracted. The repetition number calculation unit 264 calculates the extraction number (that is, three times) as the number of repetitions of the work "work E→work A→work G".

The repetition number calculation unit 264 calculates such a repetition number calculation process for each of the repetition tasks 269 identified by the repetition identification unit 256 described above. Further, the repeat count calculation unit 264 calculates the total repeat count of the repeat work 269 without distinguishing the work patterns of the repeat work. These calculated number of repetitions are output to the report generation unit 265 .

(F5. Report generator 265)
Next, functions of the report generator 265 shown in FIG. 10 will be described.

The report generation unit 265 calculates the average work time of each work content constituting the repetitive work calculated by the average time calculation unit 261, the total work time of the repetitive work calculated by the total time calculation unit 262, the number of users calculation unit A work report 270 representing the total number of users who performed repeated work calculated by 263 and the number of repetitions calculated by the repetition number calculation unit 264 is generated. Details of the work report 270 will be described later.

<G. Work report 270>
Next, a specific example of the work report 270 output from the output unit 260 (see FIG. 10) will be described with reference to FIGS. 15 to 18. FIG.

The work report 270 may be output as a paper medium after the analysis company summarizes the analysis results, may be displayed on the display 106 of the work terminal 100, or may be sent to another device capable of communicating with the work terminal 100. It may be transmitted and displayed on a display on the other device. The work report 270 may be displayed on the web browser of the work terminal 100 or the information processing terminal 200 , or may be displayed on a dedicated application installed in the work terminal 100 or the information processing terminal 200 . The work report 270 may also be generated in a format readable by various existing applications. As an example, the work report 270 may be generated as PowerPoint format data, PDF (Portable Document Format) format data, Word format data, or Excel format data. It may be generated as data in the form

(Specific example 1 of work report 270)
First, specific example 1 of the work report 270 will be described with reference to FIG. FIG. 15 is a diagram showing a work report 270A according to Concrete Example 1. As shown in FIG.

The work report 270A includes analysis result columns 271 to 276 showing analysis results of the operation log 130 collected from each of the work terminals 100. FIG.

The analysis result column 271 shows the total work time including not only the work time spent on the repeated work 269 but also the work time spent on other work. The example in FIG. 15 indicates that "2960.5 hours" were spent on all work. Further, the analysis result column 271 further indicates the total number of users subjected to analysis and the analysis target period. The total number of users is calculated based on the user identification information indicated in the operation log 130, for example.

The analysis result column 272 indicates the total work time of the repetitive work calculated by the total time calculator 262 described above. The total work time shown in the analysis result column 272 indicates the total work time of the repetitive work 269 of all patterns. The example of FIG. 15 shows that "307.5 hours" were spent on the repetitive task 269 for all patterns. Further, the analysis result column 272 further indicates the total number of users who performed the repeated task 269 and the analysis target period. The total number of users is one of the calculation results by the user number calculation unit 263 described above.

The analysis result column 273 indicates the number of repetitive work patterns identified by the analysis unit 259 described above. The example of FIG. 15 shows that 28 patterns of repetitive work have been identified.

The analysis result column 274 shows the total work time spent on personal computer work (excluding work time spent on repetitive work). The example of FIG. 15 indicates that "1165 hours" were spent on the total work time spent on personal computer work. Further, the analysis result column 274 further indicates the total number of users who have performed the repetitive task 269 and the analysis target period. The total number of users is one of the calculation results by the user number calculation unit 263 described above.

In the analysis result column 275, the breakdown of working hours is indicated by pie charts 275A and 275B. A pie chart 275A shows the total work time spent on repetitive work, the total work time spent on personal computer work (excluding the work time spent on repetitive work), and the total work time not involving personal computer work. and breakdown. Pie chart 275B corresponds to a portion of pie chart 275A. More specifically, the pie chart 275B shows the breakdown of the total work time spent on repetitive work and the total work time spent on personal computer work (excluding the work time spent on repetitive work). show.

In the analysis result column 276, the total work time for repetitive work of each work pattern is shown in a comparable form (for example, graph 276A). In the example of FIG. 15, the graph 276A is shown as a pie chart, but the graph 276A may be a bar graph, a line graph, or any other type of graph. may By showing the total work time in a comparable form for repeated work of each work pattern, the user can easily grasp which work pattern takes time.

Preferably, on the graph 276A, the total work time for each work pattern is displayed in descending order of the total work time. This makes it easier for the user to compare the total work time for each work pattern.

(Concrete example 2 of work report 270)
Next, a specific example 2 of the work report 270 will be described with reference to FIG. 16 . FIG. 16 is a diagram showing a work report 270B according to Specific Example 2. As shown in FIG.

The work report 270B includes analysis result columns 280 to 284 showing analysis results of the operation logs 130 collected from each of the work terminals 100. FIG.

The analysis result column 280 shows the average work hours per user in a predetermined period. In addition, the analysis result column 280 further indicates the average working hours per day.

The analysis result column 281 shows a predetermined number of top users with long working hours and their working hours. In the example of FIG. 16, the analysis result column 281 shows the top three users with the longest working hours and the working hours of the three users.

The analysis result column 282 shows a predetermined number of users with short working hours and their working hours. In the example of FIG. 16, the analysis result column 282 shows the bottom two users with the shortest working hours and the working hours of the two users.

The analysis result column 283 shows the breakdown of business hours for each user. The breakdown of work time is the total work time spent on repetitive work, the total work time spent on computer work (excluding work time spent on repetitive work), and the total work that does not involve computer work. Consists of time and By displaying the total work time spent on repetitive work for each user, the user can easily grasp which user is doing a lot of repetitive work.

The analysis result column 284 shows the analyst's comments on the work report 270B. Typically, the comments are entered by the analyst.

(Concrete example 3 of work report 270)
Next, a specific example 3 of the work report 270 will be described with reference to FIG. FIG. 17 is a diagram showing a work report 270C according to Specific Example 3. As shown in FIG.

The work report 270</b>C shows analysis results regarding a specific work pattern of repetitive work (hereinafter also referred to as “first repetitive work”) among the plurality of patterns of repetitive work 269 identified by the analysis unit 259 described above. . The work report 270C includes analysis result columns 310,311,313,320,330,340.

The analysis result column 310 indicates the countermeasure priority for the first repetitive work. The countermeasure priority is determined by comprehensively considering the work time spent on the first repeated work, the total number of users who performed the first repeated work, and the like. As an example, the longer the working hours, the higher the countermeasure priority. Alternatively, the greater the total number of users, the higher the countermeasure priority.

The analysis result column 311 shows the total work time spent on the first repeated work. The total work time is calculated by the total time calculator 262 described above. The example in FIG. 17 indicates that "108 hours" were spent on the first iteration. By checking the total work time, the user can easily grasp the time spent on the first repeated work.

Further, the analysis result column 311 indicates the total number of times the first repeated work has been performed. The total number of times is calculated by the repetition number calculation unit 264 . The example in FIG. 17 indicates that the first repeated work has been performed "1361 times". By checking the total number of times, the user can easily grasp the total number of times the first repeated work has been performed.

The analysis result column 313 shows the total number of users who have performed the first repetitive work, and a graph 314 showing the amount of work related to the first repetitive work in a manner that allows comparison for each user. In the example of FIG. 17, the graph 314 is shown as a pie chart, but any type of graph can be adopted as long as it allows users to compare the amount of work related to the first repetitive work. As an example, the graph 314 may employ a bar graph, a line graph, or other types of graphs. The user can easily grasp which worker is doing most of the first repetitive work in a manner in which the work amount of the first repetitive work can be compared for each user.

In the analysis result column 320, each work content constituting the first repeated work is arranged in chronological order. At this time, each work content is connected by an arrow in chronological order, for example. This allows the user to easily analyze the work order in the first repetitive work. Also, this order can serve as a work manual for the user to perform work.

Preferably, the analysis result column 320 further displays the average work time required to perform each work that constitutes the first repeated work once. Typically, each of the average task times is displayed side by side with the corresponding task content. The average work time is calculated by the above average time calculator 261 (see FIG. 10). By showing the average work time of each work content, the user can easily grasp which work content is using labor among the work content that constitutes the repetitive work, and which work content is used. It is possible to analyze whether it is effective if it is automated. In addition, the analysis result column 320 shows not only the average work time spent on each work content, but also the average work time required to perform the first repeated work once.

The analysis result column 330 shows a graph 331 that represents the frequency of the first repetitive work for each predetermined time segment. Although the graph 331 is shown as a histogram in the example of FIG. 17, any type of graph can be employed as long as it represents the temporal distribution of the frequency of the first repeated work. As an example, the graph 331 may employ a bar graph, a line graph, or other types of graphs. Also, in the example of FIG. 17, the time segments shown in the graph 331 are shown in units of days, but the time segments may be shown in units of minutes or hours. It may be indicated in units of weeks, may be indicated in units of months, or may be indicated in units of years. Also, the time unit shown in the graph 331 may be arbitrarily adjusted by the user. By checking the graph 331, the user can easily analyze when the first repetitive work is frequently performed.

Each frequency shown in the graph 331 is calculated based on the operation time shown in the above operation log 130 (or work identification result D1). Since the operation time represents the time at which the repeated work was performed, information processing terminal 200 determines whether each of the first repeated tasks falls within a plurality of predetermined time segments based on the operation time. belonging can be identified.

The analysis result column 340 shows the analyst's comments on the work report 270C. Typically, the comments are entered by the analyst.

(Concrete example 4 of work report 270)
Next, a specific example 4 of the work report 270 will be described with reference to FIG. FIG. 18 is a diagram showing a work report 270D according to Specific Example 4. As shown in FIG.

The work report 270D shows the analysis result of the repetitive work (hereinafter also referred to as "second repetitive work") having a different work pattern from the first repetitive work described above with reference to FIG. The work report 270D includes analysis result columns 410, 411, 413, 420, 430, and 440.

The analysis result column 410 indicates the countermeasure priority for the second repeated work. The countermeasure priority is determined by comprehensively considering the work time spent on the second repeated work, the total number of users who performed the second repeated work, and the like. As an example, the longer the working hours, the higher the countermeasure priority. Alternatively, the greater the total number of users, the higher the countermeasure priority.

The analysis result column 411 shows the total work time spent on the second repeated work. The total work time is calculated by the total time calculator 262 described above. The example in FIG. 18 shows that "76.7 hours" were spent on the second iteration. By checking the total work time, the user can easily grasp the time spent on the second repeated work.

In addition, the analysis result column 411 indicates the total number of times the second repeated work has been performed. The total number of times is calculated by the repetition number calculation unit 264 . The example in FIG. 18 indicates that the second iteration has been performed “955 times”. By checking the total number of times, the user can easily grasp the total number of times the second repeated work has been performed.

The analysis result column 413 shows the total number of users who performed the second repeated work, and a graph 414 showing the amount of work related to the second repeated work in a manner that allows comparison for each user. In the example of FIG. 18, the graph 414 is shown as a pie chart, but any type of graph can be adopted as long as it allows users to compare the amount of work related to the second repeated work. As an example, the graph 414 may employ a bar graph, a line graph, or other types of graphs. The user can easily grasp which worker performs the second repeated work in a manner that allows comparison of the work amount of the second repeated work for each user.

In the analysis result column 420, each work content constituting the second repeated work is arranged in chronological order. At this time, each work content is connected by an arrow in chronological order, for example. This allows the user to easily analyze the work order in the second repetitive work. Also, this order can serve as a work manual for the user to perform work.

Preferably, the analysis result column 420 further displays the average work time required to perform each work that constitutes the second repeated work once. Typically, each of the average task times is displayed side by side with the corresponding task content. The average work time is calculated by the above average time calculator 261 (see FIG. 10). By showing the average work time of each work content, the user can easily grasp which work content is using labor among the work content that constitutes the repetitive work, and which work content is used. It is possible to analyze whether it is effective if it is automated. In addition, the analysis result column 420 shows not only the average work time spent on each work content, but also the average work time required to perform the second repeated work once.

Preferably, in the analysis result column 420, each work content that constitutes the second repeated work is indicated in a different manner for each application. More specifically, each work content constituting the second repeated work is divided into a first work group performed using the first application and a second work performed using the second application. When including groups, the first task group and the second task group are shown differently. The method of classification is arbitrary. For example, the first work group and the second work group may be classified by using different colors such as characters and backgrounds, or by being surrounded by a figure such as a rectangle.

The analysis result column 430 shows a graph 431 that represents the frequency of the second repetitive work for each predetermined time segment. Although the graph 431 is shown as a histogram in the example of FIG. 18, any type of graph may be employed as long as it represents the temporal distribution of the frequency of the second repeated work. As an example, the graph 431 may employ a bar graph, a line graph, or other types of graphs. Also, in the example of FIG. 18, the time segments shown in the graph 431 are shown in units of days, but the time segments may be shown in units of minutes or hours. It may be indicated in units of weeks, may be indicated in units of months, or may be indicated in units of years. Also, the time unit shown in the graph 431 may be arbitrarily adjusted by the user. By checking the graph 431, the user can easily analyze when the second repetitive work is frequently performed.

Each frequency shown in the graph 431 is calculated based on the operation time shown in the above operation log 130 (or work identification result D1). Since the operation time represents the time at which the repeated work was performed, information processing terminal 200 determines whether each of the second repeated tasks falls within a plurality of predetermined time segments based on the operation time. belonging can be identified.

The analysis result column 440 shows the analyst's comments on the work report 270D. Typically, the comments are entered by the analyst.

<H. Flowchart >
The control structures of the work terminal 100 and the information processing terminal 200 will be sequentially described below with reference to FIGS. 19 and 20. FIG.

(H1. Control Flow of Work Terminal 100)
First, the control flow of the work terminal 100 will be described with reference to FIG. FIG. 19 is a flow chart showing a part of the processing executed by work terminal 100 . The control device 101 (see FIG. 3) of the work terminal 100 executes the process of step S110 based on receiving the instruction to start executing the operation monitoring program 122 (see FIG. 3).

In step S110, the control device 101 determines whether or not the work terminal 100 has been operated. When control device 101 determines that work terminal 100 has been operated (YES in step S110), control is switched to step S112. Otherwise (NO in step S110), control device 101 switches control to step S120.

In step S112, the control device 101 writes information related to the user's operation to the above-described operation log 130 (see FIG. 2). As an example, the control device 101 includes terminal identification information (for example, personal computer name) indicating the operation target work terminal 100, user identification information (for example, login name) indicating the user who performed the operation, and Operation time indicating the timing, working time indicating the time spent on the operation, application identification information (for example, application name) indicating the application to be operated, and operation information for the source work terminal 100 (for example, operation (the title of the target window) is written in the operation log 130 .

In step S120, the control device 101 determines whether or not the timing for transmitting the operation log 130 has come. The transmission timing of operation log 130 comes every time a predetermined period elapses (for example, one day). When control device 101 determines that the timing for transmitting operation log 130 has arrived (YES in step S120), control is switched to step S122. Otherwise (NO in step S120), control device 101 switches control to step S130.

In step S<b>122 , the control device 101 transmits the operation log 130 to the information processing terminal 200 .

In step S130, the control device 101 determines whether or not to terminate the execution of the operation monitoring program 122. As an example, the control device 101 determines to end the execution of the operation monitoring program 122 based on receiving an operation to end the execution of the operation monitoring program 122 from the user. When control device 101 determines to end the execution of operation monitoring program 122 (YES in step S130), it ends the process shown in FIG. Otherwise (NO in step S130), control device 101 returns the control to step S110.

(H2. Control flow of information processing terminal 200)
Next, with reference to FIG. 20, the control flow of information processing terminal 200 will be described. FIG. 20 is a flow chart showing part of the process executed by information processing terminal 200 . The control device 201 (see FIG. 4) of the information processing terminal 200 starts executing the process of step S210 based on receiving the instruction to start executing the information processing program 222 (see FIG. 4).

In step S<b>210 , the control device 201 functions as the communication unit 252 (see FIG. 2 ) and receives the operation log 130 from each of the work terminals 100 .

In step S<b>212 , the control device 201 functions as the above-described work identification unit 254 (see FIG. 2 ), and from time-series operation information defined in each operation log 130 received from the work terminal 100 , each operation For information, identify the corresponding work content. The method of identifying the work content corresponding to the operation information is as described in the above-mentioned "B. Outline of information processing terminal 200", so description thereof will not be repeated.

In step S214, the control device 201 functions as the above-described repetition identification unit 256 (see FIG. 2), and from each work content identified for each piece of time-series operation information in step S212, a series of repetitions that are being repeatedly performed are identified. Identify work. Since the method of specifying the repeated work is as described in the above-mentioned "E. Functions of the repeated work specifying unit 256", the description thereof will not be repeated.

In step S216, the control device 201 functions as the above-described output unit 260 (see FIG. 2), and analyzes the operation log 130 based on the work pattern of repetitive work identified in step S214. The method of analyzing operation log 130 is as described above in "F. Function of output unit 260", so description thereof will not be repeated.

In step S<b>218 , the control device 201 functions as the output unit 260 (see FIG. 2 ) described above, and outputs the analysis result in step S<b>216 as the work report 270 . The work report 270 has been described in "G. Work report 270", so the description thereof will not be repeated.

<I. Modification 1>
With reference to FIG. 21, a modified example of the learning data set generation method by the learning data generation unit 257 (see FIG. 7) will be described.

In the above description, the learning data generation unit 257 blocks the time-series work content group included in the work identification result D1 by a predetermined number of units, and generates the learning data set D2 based on each block. On the other hand, in this modification, the learning data generation unit 257 blocks the work content group defined in time series in the work identification result D1 for each predetermined time unit, and based on each block, the learning data set D2A. to generate

FIG. 21 is a diagram schematically showing the function of learning data generator 257 according to the modification. FIG. 21 shows a work content group included in the work identification result D1 as a graph G1. The horizontal axis of graph G1 represents time. The vertical axis of the graph G1 represents the type of work content. Each point shown in the graph G1 indicates one work content included in the work identification result D1.

As shown in FIG. 21, the learning data generator 257 blocks the work content group included in the work identification result D1 for each predetermined unit time ΔT. Next, the learning data generator 257 interpolates between points included in each block, and normalizes the work content group included in the unit time ΔT to a predetermined number of dimensions. Any interpolation method may be used. As an example of interpolation processing, the points may be connected by straight lines, or an approximation function of the points may be calculated. This absorbs the difference in the number of work contents included in each block. Below, the work content group after normalization is also called a "work vector."

Next, the learning data generation unit 257 tags the working vector of each block with the working vector corresponding to the next block following the block as correct information. In the example of FIG. 21, for example, the work vector of work pattern 4 following work pattern 1 is tagged as correct information to the work vector corresponding to work pattern 1 . Further, the work vector of work pattern 5 following work pattern 2 is tagged as correct information to the work vector corresponding to work pattern 2 . Such tagging is performed for all blocks and they are generated as the training data set D2A.

The learning unit 258 (see FIG. 8) described above generates the work estimation model M based on the generated learning data set D2A. Since the method of generating the task estimation model M is as described above, the description thereof will not be repeated.

The analysis unit 259 (see FIG. 9) described above inputs the learning data set D2A again to the work estimation model M that has been learned by the learning unit 258, and identifies work patterns indicating repetitive work.

In this way, the task estimation model M is generated based on the learning data set D2A in which the length of time is taken into account, so that repetitive tasks can be extracted more reliably.

<J. Modification 2>
Next, a modified example of a method of specifying repeated work by the repeat specifying unit 256 (see FIG. 2) will be described.

Although the repetitive identification unit 256 described above identifies the repetitive work 269 based on the work estimation model M generated by the learning process, the repetitive work may be identified by a method other than the learning process.

As an example, the repetition identification unit 256 blocks the time-series work content group included in the work identification result D1 by a predetermined number of units (or by a predetermined time unit), and compares each block of the work content group. Calculate the similarity between blocks. More specifically, the repetition identification unit 256 regards the work content group included in each block as a vector, and calculates the inner product value of the vector between each block. Blocks with high inner product values indicate that the work contents are similar.

As described above, the repetition identification unit 256 may identify the repetition work 269 through learning processing, or may identify the repetition work 269 through predetermined rule-based processing.

<K. Modification 3>
With reference to FIG. 22, a modification of the implementation form of the functional configuration will be described.

In the example of FIG. 2 described above, functional configurations such as the work identification unit 254 , the work identification unit 254 , the repetition identification unit 256 , and the output unit 260 are implemented in the information processing terminal 200 . That is, various functions for generating the work report 270 from the operation log 130 were implemented in the information processing terminal 200 . On the other hand, in this modified example, these functional configurations are implemented in the work terminal 100 .

FIG. 22 is a diagram showing a work terminal 100 according to a modification. As shown in FIG. 22, the work terminal 100 includes a control device 101 and a storage device 120 as a hardware configuration. The control device 101 includes an operation monitoring unit 152, a work identification unit 254, a repetition identification unit 256, and an output unit 260 as functional configurations.

Since the functions of work identification unit 254, work identification unit 254, repetition identification unit 256, and output unit 260 are as described above, description thereof will not be repeated below.

The operation monitoring unit 152 monitors operations on the work terminal 100 . The operation monitoring unit 152 writes information related to the user's operation to the above operation log 130 (see FIG. 2) based on the operation of the work terminal 100 by the user. As an example, the operation monitoring unit 152 uses terminal identification information (for example, personal computer name) indicating the work terminal 100 to be operated, user identification information (for example, login name) indicating the user who performed the operation, and operation time indicating the timing of the operation, working time indicating the time spent on the operation, application identification information (for example, application name) indicating the application to be operated, and operation information for the source work terminal 100 (for example, title of the window to be operated) is written in the operation log 130.

As described above, various processes for generating the work report 270 from the operation log 130 do not necessarily have to be implemented in the information processing terminal 200 . These various processes may be implemented in the work terminal 100, for example. That is, the work terminal 100 can function as the information processing terminal 200 described above.

<L. Summary>
As described above, based on the operation log 130 received from each of the work terminals 100, the information processing terminal 200 identifies work that is being performed repeatedly (that is, repeated work). The information processing terminal 200 then outputs the identified repetitive work as the work report 270 .

If repetitive work is automated, work efficiency will be greatly improved, so presentation of such repetitive work can be a hint for improving work efficiency. Therefore, the information processing terminal 200 can help improve the work efficiency of the work terminal 100 .

The information processing terminal 200 not only presents repetitive work, but also outputs various hints for improving work as a work report 270 . As an example, the information processing terminal 200 outputs various information such as the total number of repeated tasks, a breakdown of the total number of times for each user, and the average work time of each task that constitutes the repeated tasks. These pieces of information can also be hints for improving work efficiency. By presenting these pieces of information, the user can improve work efficiency more efficiently.

It should be considered that the embodiments disclosed this time are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the scope of the claims rather than the above description, and is intended to include all modifications within the meaning and range of equivalents of the scope of the claims.

100 work terminal, 101, 201 control device, 102, 202 ROM, 103, 203 RAM, 104, 204 communication interface, 105, 205 display interface, 106, 206 display, 107, 207 input interface, 108, 208 input device, 110 , 210 bus, 120, 220 storage device, 122 operation monitoring program, 130 operation log, 132 work program, 152 operation monitoring unit, 200 information processing terminal, 222 information processing program, 232 work definition, 252 communication unit, 254 work identification 256 repetition identification unit 257 learning data generation unit 258 learning unit 259 analysis unit 260 output unit 261 average time calculation unit 262 total time calculation unit 263 number of users calculation unit 264 number of times calculation unit 265 report Generation unit 269 Repetitive work 270, 270A, 270B, 270C, 270D Work report 271, 272, 273, 274, 275, 276, 280, 281, 282, 283, 284, 310, 311, 313, 320, 330 , 340, 410, 411, 413, 420, 430, 440 analysis result column, 275A, 275B pie chart, 276A, 314, 331, 414, 431 graph, 500 information processing system.

Claims (10)

A communication unit that communicates with a plurality of business terminals, wherein the communication unit receives an operation log of the business terminal from each of the plurality of business terminals, and the operation log includes operation information for the business terminal that is the transmission source. Each is specified in chronological order, and furthermore,
a work identification unit for identifying work content corresponding to each piece of operation information from each piece of time-series operation information defined in each received operation log;
Of the work patterns extracted from each work content specified for each piece of time-series operation information by the work specifying unit, a work pattern whose correlation value with respect to a predetermined work pattern exceeds a predetermined value is repeatedly performed. a repeat identification unit for identifying as a series of repetitive tasks that
An information processing terminal, comprising: an output unit for arranging and outputting work contents constituting the repetitive work in chronological order. Each piece of operation information defined in each received operation log is further associated with a working time spent on the operation,
The output unit calculates an average work time spent on the work content for each work content that constitutes the repetitive work based on each work time specified in each of the received operation logs, and 2. The information processing terminal according to claim 1, further outputting each of the calculated average work times. 3. The output unit according to claim 2, wherein the output unit calculates a total work time spent on the repeated work based on each work time defined in each of the received operation logs, and further outputs the total work time. Information processing terminal described. When a plurality of patterns of repetitive work are identified by the repetition identification unit, the output unit calculates the total work time for each of the plurality of patterns of repetitive work, and calculates the total work time for each of the plurality of patterns of repetitive work. 4. The information processing terminal according to claim 3, further outputting a graph showing work times in a comparable manner. Each piece of operation information defined in each received operation log is further associated with user identification information indicating a user who performed the operation,
5. The output unit according to any one of claims 1 to 4, wherein the output unit calculates a total number of users who have performed the repeated work based on the user identification information defined in each of the received operation logs, and further outputs the total number. or the information processing terminal according to item 1. 6. The information processing terminal according to claim 5, wherein said output unit further outputs a graph representing the amount of work related to said repetitive work in a comparable manner for each user. The output unit identifies the number of times the repeated work is included in each work content identified for each piece of time-series operation information by the work identification unit, and determines the number of times the repeated work is performed. The information processing terminal according to any one of claims 1 to 6, further outputting as the number of times. Each piece of operation information defined in each received operation log is further associated with an operation time at which the operation was performed,
The output unit calculates the frequency of the repeated work for each predetermined time segment based on each operation time specified in each of the received operation logs, and calculates for each predetermined time segment 8. The information processing terminal according to any one of claims 1 to 7, further outputting a graph showing the frequency in the . A computer-executed information processing method comprising:
a step of receiving an operation log of the business terminal from each of the plurality of business terminals, wherein the operation log defines each piece of operation information for the business terminal of the transmission source in chronological order;
a step of identifying work content corresponding to each piece of operation information from each piece of time-series operation information defined in each of the received operation logs;
Of the work patterns extracted from each work content identified for each piece of time-series operation information in the identifying step, those whose correlation value with respect to a predetermined work pattern exceeds a predetermined value are repeatedly performed. a step identified as a series of repetitive tasks that
an information processing method comprising the step of arranging and outputting each work content constituting the repetitive work in chronological order. An information processing program executed by a computer,
The information processing program causes the computer to:
executing a step of receiving an operation log of the business terminal from each of the plurality of business terminals, wherein the operation log defines each piece of operation information for the business terminal of the transmission source in chronological order;
The information processing program further causes the computer to:
a step of identifying work content corresponding to each piece of operation information from each piece of time-series operation information defined in each of the received operation logs;
Of the work patterns extracted from each work content identified for each piece of time-series operation information in the identifying step, those whose correlation value with respect to a predetermined work pattern exceeds a predetermined value are repeatedly performed. a step identified as a series of repetitive tasks that
and a step of arranging and outputting each work content constituting the repetitive work in chronological order.

Images