JP6048254B2 - Operation support program, operation support method, and information processing apparatus - Google Patents

Description

  The present invention relates to an operation support program.

  Conventionally, recommendation technology has been used in various fields such as recommendation of a web page. For example, in the recommendation technique, interests and interests on the web are collected for each user, and information on the obtained interests and interests having a high score is provided to another user.

  As an example of the recommendation technique, the recommendation device learns a prediction model in consideration of the order of items in which a plurality of users have taken action and the relationship between items expressed by the degree of link density. Then, the recommendation device predicts an item on which the user next takes an action using the learned prediction model. Further, the recommendation device determines an item to be recommended to the user based on the predicted item. Items include web pages and the like.

JP 2011-154591 A JP 2008-513887 A JP 2001-5862 A JP 2005-33600 A

  However, the conventional recommending technique has a problem that a series of user operations cannot be recommended. That is, in the conventional recommendation device, the prediction target is a web page, and the web page itself can be recommended, but the transition of the web page cannot be recommended. That is, the recommendation device cannot recommend a series of user operations.

  An object of one aspect is to be able to recommend a series of user operations.

  The operation support program disclosed in the present application receives an evaluation request for an operation candidate that can be operated next for a predetermined node in the operation sequence being operated, and evaluates each user for each operation for each node in the operation sequence. Processing for calculating the evaluation value of the next operation candidate that can be operated for the node that has received the operation evaluation request based on the first information that is indicated and the second information that indicates the distance between the users. Each of the evaluation values calculated by the above is executed in association with each operation candidate and transmitted to the transmission source.

  According to one aspect, it is possible to recommend a series of user operations.

FIG. 1 is a functional block diagram illustrating the configuration of the operation support system according to the embodiment. FIG. 2 is a diagram illustrating an example of a data structure of the operation sequence DB. FIG. 3 is a diagram illustrating an example of an operation sequence. FIG. 4 is a diagram illustrating an example of the data structure of the action DB. FIG. 5 is a diagram illustrating an example of the data structure of the recommendation DB. FIG. 6 is a diagram illustrating an example of a transition graph. FIG. 7 is a diagram illustrating an example of the data structure of the normalized recommendation DB. FIG. 8 is a diagram illustrating an example of a social graph. FIG. 9 is a diagram illustrating an example of the data structure of the social graph DB. FIG. 10A is a diagram (1) illustrating a transition graph and an evaluation value. FIG. 10B is a diagram (2) illustrating a transition graph and an evaluation value. FIG. 11 is a flowchart of the operation sequence recording process on the PC side. FIG. 12 is a flowchart of the operation record request process on the PC side. FIG. 13A is a diagram (1) illustrating a flowchart of a recommendation display process on the PC side. FIG. 13B is a diagram (2) illustrating a flowchart of a recommendation display process on the PC side. FIG. 13C is a flowchart of the page reading / displaying process. FIG. 14 is a flowchart of the operation recording process on the recommendation recording server side. FIG. 15 is a diagram illustrating a flowchart of a recommendation calculation process on the recommendation transmission server side. FIG. 16A is a diagram (1) illustrating an example of a screen. FIG. 16B is a diagram (2) illustrating an example of the screen. FIG. 16C is a diagram (3) illustrating an example of the screen. FIG. 16D is a diagram (4) illustrating an example of the screen. FIG. 16E is a diagram (5) illustrating an example of the screen. FIG. 16F is a diagram (6) illustrating an example of the screen. FIG. 17A is a diagram (1) illustrating an example of recommendation display. FIG. 17B is a diagram (2) illustrating an example of recommendation display. FIG. 18 is a diagram illustrating an example of a computer that executes an operation support program.

  Embodiments of an operation support program, an operation support method, and an information processing apparatus disclosed in the present application will be described below in detail with reference to the drawings. The present invention is not limited to the embodiments.

[Configuration of Operation Support System According to Embodiment]
FIG. 1 is a block diagram illustrating the configuration of the operation support system according to the embodiment. As shown in FIG. 1, the operation support system 9 includes a PC (Personal Computer) 1, a PC 2, a recommendation recording server 3, a recommendation transmission server 4, and a web server 5. The PC 1 and the PC 2 are connected to the web server 5 via the Internet 6. The web server 5 implements a web application as a web service. In the embodiment, the web application is described as a web application related to the “patent application system”. The network that the PC 1 and PC 2 connect to the web server 5 is not limited to the Internet 6, and may be an in-house LAN (Local Area Network) or an in-house intranet.

  The operation support system 9 aggregates the operation sequence of the browser operated for each user together with the evaluation of the operation determined for each user. And the operation assistance system 9 recommends the operation candidate which should be operated next based on the collected evaluation and the distance between users at the time of operation of the user's browser operated later. The “operation sequence” here refers to a series of browser operations. “Operation” means an “action” executed by a user from a browser page, and examples of the operation element include links and form transmission. Form transmission includes, for example, a text box, radio button, check box, and file selection. Hereinafter, such an operation support system 9 will be described in detail.

  The PC 1 records the operation sequence of the browser operated by the user and the evaluation of the operation, and transmits the recorded result to the recommendation recording server 3 described later. The PC 1 has a browser 11 and an operation sequence DB (Database) 12. Further, the browser 11 includes an operation sequence recording unit 111 and an operation record requesting unit 112. The PC 1 corresponds to, for example, a personal computer, but may be any communication device that can be operated, such as a mobile phone or a smartphone.

  The operation sequence recording unit 111 records an operation sequence. For example, the operation sequence recording unit 111 records the action executed by the user from the page displayed by the browser 11 and the URL (Uniform Resource Locator) of the destination page of the action in the operation sequence DB 12.

  Here, the data structure of the operation sequence DB 12 will be described with reference to FIG. FIG. 2 is a diagram illustrating an example of a data structure of the operation sequence DB. As shown in FIG. 2, the operation sequence DB 12 stores a tab ID (identification) 12a, a URL 12b, an action 12c, and an evaluation 12d in association with each other. The tab ID 12a is an identifier attached to each operation sequence. The URL 12b is a URL indicating a page. The action 12c is an action executed from the page indicated by the URL 12b, that is, an operation. The action 12c stores an operation element corresponding to the executed operation, for example, a link name or a form name. Note that the name of the link and the name of the form are set by JavaScript (registered trademark), for example. The evaluation 12d indicates evaluation of a series of operations up to the page indicated by the URL 12b. In the evaluation 12d, “OK” indicating that the evaluation of the series of operations is good, “NG” indicating that the evaluation of the series of operations is bad, or blank indicating that neither is stored. The evaluation 12d is stored when, for example, a “good example” button and a “bad example” button added to the page by the function of the browser 11 are pressed by the user.

  As an example, when the tab ID 12a is “1”, “top.html” is stored as the URL 12b, “link 1” is stored as the action 12c, and “NG” is stored as the evaluation 12d. That is, it is stored that the operation of “link 1” is executed in “top.html” and a series of operations so far are evaluated as “NG”. In the second line, “inventor.html” is stored as URL 12b, “form 1” is stored as action 12c, and blank is stored as evaluation 12d. That is, it is stored that the operation of “form 1” is executed on the page “inventor.html” which is the transition destination of the action “link 1” executed by the user from “top.html”. In the third line, “idea.html” is stored as the URL 12b, “form 2” is stored as the action 12c, and “OK” is stored as the evaluation 12d. In the fourth line, “save.html” is stored as the URL 12b, blank as the action 12c, and blank as the evaluation 12d.

  Returning to FIG. 1, the operation record request unit 112 requests the recomment recording server 3 to record the operation sequence. For example, when a series of operations is completed, the operation record request unit 112 determines whether or not there is at least one “OK” or “NG” in the operation sequence evaluation 12d stored in the operation sequence DB 12. To do. If the operation record requesting unit 112 determines that there is at least one “OK” or “NG” in the operation sequence evaluation 12d, the operation record request unit 112 stores information from the operation sequence DB 12 to the last recorded “OK” or “NG”. Extract the operation sequence. The extracted operation sequence is composed of a pair of URL 12b and action 12c up to the last recorded “OK” or “NG”. Then, the operation record request unit 112 transmits the extracted operation sequence, the user name, and the last recorded “OK” or “NG” to the recommendation recording server 3.

  Here, the operation sequence will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of an operation sequence. Each page will be described as a node. FIG. 3 shows an operation sequence in which the user A executes a series of operations sequentially from the node “top.html”. Here, for example, the user A executes the action of the form 1 at the node “top.html”, and transitions to the node “idea.html”. The lower part of FIG. 3 shows an operation sequence in which the user E executes a series of operations in order from the node “top.html”. Here, for example, the user E executes the action of the form 1 at the node “top.html” and transitions to the node “idea.html”.

  It is assumed that the user A presses a “good example” button at the node “complete.html” in such an operation sequence. Then, the operation record request unit 112 transmits the operation sequence of the user A, the user name “A”, and “OK” to the recommendation recording server 3. Further, it is assumed that the user E presses the “bad example” button at the node “error.html”. Then, the operation record request unit 112 transmits the operation sequence of the user E, the user name “E”, and “NG” to the recommendation recording server 3.

  Returning to FIG. 1, the PC 2 transmits an evaluation request for the next operable operation candidate to the recommended transmission server 4 for a predetermined page of the operation sequence being operated, and adds the evaluation result to the page. The PC 2 has a browser 21 and an action DB 22. Further, the browser 21 includes a recommendation request unit 211 and a recommendation display unit 212. The PC 2 corresponds to, for example, a personal computer, but may be any communication device that can be operated, such as a mobile phone or a smartphone.

  The recommendation request unit 211 displays a page with a predetermined URL while operating the web application. Then, the recommendation request unit 211 transmits an evaluation request to the recommendation transmission server 4 for an operation candidate (action) that can be operated next from the displayed page. For example, the recommendation request unit 211 reads the page of the current URL from the web server 5 for the operation sequence being operated, and displays the read page. At this time, the recommendation request unit 211 extracts an operation (action) element from the read page and records it in the action DB 22. Then, the recommendation request unit 211 retrieves all the actions that can be operated next from the action DB 22, and uses the extracted actions, the operation sequence so far, and the user name of the operating user as an evaluation request. Include and send.

  The recommendation display unit 212 receives evaluation values for all actions from the recommendation transmission server 4, and adds the received evaluation values to the displayed page for display. For example, the recommendation display unit 212 displays each evaluation value in correspondence with all actions that can be operated next. An example of the evaluation value will be described later.

  Here, the data structure of the action DB 22 will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of the data structure of the action DB. As shown in FIG. 4, the action DB 22 stores a tab ID 22a, a URL 22b, and an action 23c in association with each other. The tab ID 22a is an identifier attached to each operation sequence. URL 22b is the URL of the currently displayed page. The action 23c is an action of the currently displayed page. That is, the action 23c stores the next operable action for the currently displayed page.

  As an example, when the tab ID 22a is “1”, “idea.html” is stored as the URL 22b, and “form 1”, “form 2”, and “link 1” are stored as the action 22c.

  Returning to FIG. 1, the recommendation recording server 3 includes a control unit 31 and a recommendation DB 32. The control unit 31 includes an operation recording unit 311. The control unit 31 is, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array) or an electronic circuit such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). The recommendation DB 32 stores an evaluation of each user for each operation on each page of the operation sequence. The data structure of the recommendation DB 32 will be described later.

  The operation recording unit 311 records the evaluation of each user's operation sequence in the recommendation DB 32. For example, when receiving the operation sequence, the user name, and the last evaluation of the operation sequence from the PC 1, the operation recording unit 311 compares the received operation sequence with the recommendation DB 32. If there is a new page (node) or a new transition destination (path) in the received operation sequence, the operation recording unit 311 adds it to the recommendation DB 32. Then, the operation recording unit 311 adds a user name to a location corresponding to the received evaluation in the recommendation DB 32 for all pages that match the operation sequence. Accordingly, the operation recording unit 311 can generate a transition graph using the recommendation DB 32.

  Here, the data structure of the recommendation DB 32 will be described with reference to FIG. FIG. 5 is a diagram illustrating an example of the data structure of the recommendation DB. As shown in FIG. 5, the recommendation DB 32 stores ID 32a, URL 32b, OK 32c, NG 32d, action 32e, and transition destination ID 32f in association with each other. The ID 32a is an identifier attached to each URL. URL 32b is the URL of the page. OK 32c means a list of users who pressed the “good example” button for each URL (node). NG32d means a list of users who pressed the “bad example” button for each URL (node). The action 32e is an action executed from the page indicated by the URL 32b, that is, an operation. The action 32e stores an operation element corresponding to the executed operation, for example, a link name or a form name. The transition destination ID 32f means the ID of the page (node) transitioned by the action 32e, in other words, a path.

  As an example, when the ID 32a is “1”, “top.html” is stored as the URL 32b, “A, C, D, I” is stored as the OK 32c, and “B, E, F, H, J” is stored as the NG 32d. . As the action 32e, “link 1” is stored in the first line, and “link 2” is stored in the second line. As the transition destination ID 32f, “2” is stored in the first row and “11” is stored in the second row.

  Here, an example of a transition graph using the recommendation DB 32 will be described with reference to FIG. FIG. 6 is a diagram illustrating an example of a transition graph. FIG. 6 is a transition graph using the recommendation DB 32 shown in FIG. As shown in FIG. 6, in the transition graph, for each node, the user list of the user who pressed the “good example” button and the user list of the user who pressed the “bad example” button are set as additional information. The Here, in the node “top.html”, “A, C, D, I” is set in the user list (OK side) of the user who pressed the “good example” button, and the “bad example” button is pressed. “B, E, F, H, J” is set in the user list (NG side) of the selected user. In the node “inventor.html”, “A, D, I” is set in the user list (OK side) of the user who pressed the “good example” button, and the user who pressed the “bad example” button. “B, E” is set in the user list (NG side). The additional information may further include a user list of users who have browsed.

  Returning to FIG. 1, the recommendation transmission server 4 includes a control unit 41, a recommendation DB (normalization) 42, and a social graph DB 43. The control unit 41 includes a recommendation calculation unit 411. The control unit 41 is, for example, an integrated circuit such as an ASIC or FPGA or an electronic circuit such as a CPU or MPU. The recommendation DB (normalization) 42 includes records used when the recommendation is calculated among the records stored in the recommendation DB 32. Social graph DB43 shows the distance between each user. The data structures of the recommendation DB (normalization) 42 and the social graph DB 43 will be described later.

  The recommendation calculation unit 411 receives from the PC 2 an evaluation request for an operation candidate that can be operated next for a predetermined page of the operation sequence being operated. Then, the recommendation calculation unit 411 calculates an evaluation value of an operation candidate that can be operated next for each page for which an operation evaluation request has been received, based on the recommendation DB (normalization) 42 and the social graph DB 43. Then, the recommendation calculating unit 411 associates each calculated evaluation value with each operation candidate, and transmits it to the transmission source.

  For example, the recommendation calculation unit 411 receives all the actions that can be operated next for the operation sequence being operated and the evaluation target page of the operation sequence. Then, the recommendation calculation unit 411 determines whether there is a path that matches the received operation sequence in the transition graph. Then, if there is a path that matches the received operation sequence, the recommendation calculation unit 411 determines whether there is an action that matches any of the received actions on the evaluation target page in the transition graph. Then, if there is an action that matches any of the received actions, the recommendation calculation unit 411 acquires a recommendation DB (normalization) 42 obtained by normalizing the recommendation DB 32 from the recommendation recording server 31 for the page to be evaluated. . Then, the recommendation calculation unit 411 calculates an evaluation value for each action that matches the evaluation target page in the transition graph based on the recommendation DB (normalization) 42 and the social graph DB 43.

As an example, the recommendation calculation unit 411 performs an evaluation value that becomes a “good example” and an evaluation that becomes a “bad example” when an action that can be operated next is performed based on the recommendation DB (normalization) 42 and the social graph DB 43. Calculate the value. That is, the recommendation calculation unit 411 indicates how many users having a strong relationship with the user who requested the evaluation are included in the social graph DB 43 and the user list of the transition destination when the action is performed. The evaluation value is calculated using The evaluation function is defined as the following formula (1). Note that me is an identifier representing a user who has made an evaluation request. L represents a user list. The length (me, p) indicates the strength of the relationship between me and the user p, and is the shortest distance on the social graph DB 43.

That is, the recommendation calculation unit 411 uses relation (me,) for the transition destination node u = (i, a) when the action a is executed from the i-th node on the transition graph, using Expression (1). L _OK (u)) and relation (me, L _NG (u)) are calculated. L _OK (u) is a user list on the OK side at the node u. L _NG (u) is a user list on the NG side at the node u. Hereinafter, relation (me, L _OK (u)) is N _OK (i, a), and relation (me, L _NG (u)) is N _NG (i, a).

Then, the recommendation calculation unit 411 calculates the probability of N _OK (i, a) when the action a is executed from the i-th node as an evaluation value for the action a. The evaluation formula for obtaining the evaluation value is defined as the following formula (2). Note that n is the maximum number of nodes in the operation sequence up to the page (node) to be evaluated.

Such an evaluation formula is referred to as a “recommendation evaluation formula”. In the “recommendation evaluation formula”, the recommendation calculation unit 411 sets the weight as in the following formula (3) so as to highly evaluate the recommendation of either the backward operation or the latest operation. Also good.

  Here, f (i) in Expression (3) is a set weight. When it is desired to highly evaluate the recommendation for the backward operation, for example, f (i) = i may be set. On the other hand, when it is desired to highly evaluate a recommendation for the most recent operation, for example, f (i) = 1 / i may be set.

Then, the recommendation calculation unit 411 transmits the evaluation value for the action a from the evaluation target node to the transmission source PC 2. In the PC 2, the recommendation display unit 212 displays the evaluation value in association with the operation element that can be operated next. Note that the recommendation calculation unit 411 may transmit N _OK (i, a) and N _NG (i, a) as evaluation values to the transmission source PC 2.

  Here, the data structure of the recommendation DB (normalization) 42 will be described with reference to FIG. FIG. 7 is a diagram illustrating an example of a data structure of the recommendation DB (normalization). FIG. 7 shows a recommendation DB (normalized) 42 normalized with respect to “top.html” assuming that the evaluation target page is the URL “top.html”.

As shown in FIG. 7, the recommendation DB (normalization) 42 stores URL 42a, action 42b, _OK 42c, _NG 42d, N _OK 42e, and N _NG 42f in association with each other. The URL 42a is a URL to which an action is transitioned. That is, the URL 42a corresponds to u of the transition destination node u = (i, a) described above. The action 42b indicates an action when transitioning to the page indicated by the URL 42a. That is, the action 42b corresponds to “a” of the transition destination node u = (i, a) described above. OK 42c means a list of users who have pressed the “good example” button for the page indicated by the URL 42a. OK42c corresponds to L _OK (u) described above. NG42d means a list of users who pressed the “bad example” button for the page indicated by the URL 42a. The _NG 42d corresponds to the above-described L _NG (u). N _OK 42e is a good example evaluation value for the transition destination node u = (i, a). N _OK 42e corresponds to N _OK (i, a). N _NG 42f is a bad example evaluation value for the transition destination node u = (i, a). N _NG 42f corresponds to N _NG (i, a).

As an example, when the URL 42a is “idea.html”, the action 42b is “link 1”, the OK 42c is “A, D, I”, the NG 42d is “B, E”, the N _OK 42e is “1”, N _NG 42 is stored as “1”. When the URL 42a is “inventor.html”, the action 42b is “link 2”, the OK 42c is “C”, the NG 42d is “F, H, J”, the N _OK 42e is “1”, and the N _NG 42f is “0. 833333 ".

  Next, an example of a social graph will be described with reference to FIG. FIG. 8 is a diagram illustrating an example of a social graph. As shown in FIG. 8, the distance between each user is represented. When the users are connected, the shortest distance is 1. For example, users with me and distance 1 are A, B, and C. Users with me and distance 2 are D, E, and F. Users with me and distance 3 are G and H.

  The data structure of the social graph DB 43 in such a social graph will be described with reference to FIG. FIG. 9 is a diagram illustrating an example of the data structure of the social graph DB 43. As shown in FIG. 9, the social graph DB 43 stores a user name 43a and an adjacent user name 43b in association with each other. The user name 43a is a user name. The adjacent user name 43b is a name of a user adjacent to the user indicated by the user name 43a. That is, the adjacent user name 43b indicates a user whose distance is 1 from the user indicated by the user name 43a. Note that the social graph DB 43 is stored in advance.

[Transition graph and evaluation value]
Next, the evaluation value calculated by the recommendation calculation unit 411 will be described in association with the transition graph. 10A and 10B are diagrams showing evaluation values calculated according to the transition graph shown in FIG. FIG. 10A associates “number of people” (N _OK (i, a) + N _NG (i, a) out of N _OK (i, a)) for each action as an evaluation value. FIG. 10B associates “probability” calculated by Expression (3) as an evaluation value for each action.

  As illustrated in FIG. 10A, for example, the recommendation calculation unit 411 evaluates an operation candidate that can be operated next for the first node of the operation sequence. The recommendation calculation unit 411 associates “1.5 out of 3 people” with the action for the second node as an evaluation value. Also, the recommendation calculating unit 411 associates “1 out of 1.8” with the action for the 11th node as an evaluation value.

  Also, for example, the recommendation calculation unit 411 evaluates an operation candidate that can be operated next for the third node in the operation sequence. The recommendation calculation unit 411 associates “0 out of 2.5” with the action for the fourth node as an evaluation value. Also, the recommendation calculation unit 411 associates “1.5 out of 3” with the action for the ninth node as an evaluation value.

  As illustrated in FIG. 10B, for example, the recommendation calculation unit 411 evaluates an operation candidate that can be operated next for the first node of the operation sequence. The recommendation calculation unit 411 associates “50%” as an evaluation value with the action for the second node. Also, the recommendation calculation unit 411 associates “55%” as an evaluation value with the action for the eleventh node.

  Also, for example, the recommendation calculation unit 411 evaluates an operation candidate that can be operated next for the third node in the operation sequence. The recommendation calculation unit 411 associates “60%” with the action for the fourth node as an evaluation value. Also, the recommendation calculation unit 411 associates “0%” with the action for the ninth node as an evaluation value.

[Procedure of operation sequence recording process on PC 1 side]
Next, the procedure of the operation sequence recording process on the PC 1 side will be described with reference to FIG. FIG. 11 is a flowchart of the operation sequence recording process on the PC side. In the PC 1, it is assumed that the browser 11 is activated by the user and a page is displayed.

  First, the operation sequence recording unit 111 determines whether an action has been executed from the displayed page (step S11). For example, the operation sequence recording unit 111 determines whether or not an operation element is clicked by the user. As an example, whether or not an operation element is clicked has an event attribute of “onClick” in JavaScript.

  When it is determined that the action has not been executed (step S11; No), the operation sequence recording unit 111 repeats the determination process until the action is executed. On the other hand, when it is determined that the action has been executed (step S11; Yes), the operation sequence recording unit 111 adds the executed action to the operation sequence DB 12 (step S12).

  Then, the operation sequence recording unit 111 adds the action transition destination URL to the operation sequence DB 12 (step S13). Then, the operation sequence recording process ends for the executed action.

[Procedure of operation record request processing on the PC 1 side]
Next, the procedure of the operation record request process on the PC 1 side will be described with reference to FIG. FIG. 12 is a flowchart of the operation record request process on the PC side. In the PC 1, it is assumed that the browser 11 is activated by the user and a page is displayed.

  First, the operation record request unit 112 determines whether a series of operations has been completed (step S21). For example, the operation record requesting unit 112 determines whether or not a completion button that is displayed on a page and is pressed when a series of operations is finished is pressed. When it is determined that the series of operations has not been completed (step S21; No), the operation record request unit 112 repeats the determination process until the series of operations is completed.

  On the other hand, when it is determined that a series of operations has been completed (step S21; Yes), the operation record request unit 112 determines whether or not OK / NG has been pressed once or more (step S22). For example, the operation record request unit 112 determines whether or not there is at least one “OK” or “NG” in the operation sequence evaluation 12 d stored in the operation sequence DB 12.

  When it is determined that OK / NG has not been pressed once or more (step S22; No), the operation record request unit 112 ends the operation record request process. On the other hand, when it is determined that OK / NG is pressed once or more (step S22; Yes), the operation record request unit 112 extracts the operation sequence recorded in the operation sequence DB 12 (step S23). For example, the operation record request unit 112 cuts the operation sequence recorded in the operation sequence DB 12 at a portion corresponding to the last OK / NG, and extracts the operation sequence of the cut first half portion.

  Then, the operation record requesting unit 112 transmits the operation sequence of the cut first half part, the user name and the last “OK” or “NG” (evaluation) to the recommendation recording server 3 (step S24). Then, the operation record request unit 112 ends the operation record request process. Thereafter, the browser 11 is terminated by the user.

[Procedure for recommendation display processing on PC2 side]
Next, with reference to FIG. 13A to FIG. 13C, a procedure of recommendation display processing on the PC 2 side will be described. FIG. 13A is a diagram illustrating a flowchart of a recommendation display process on the PC side in which the recommendation is displayed by the number of people. In the PC 2, it is assumed that the browser 11 is activated by the user.

  The recommendation request unit 211 executes a page reading / display process for a URL (step S31). The page reading / displaying process will be described with reference to FIG. 13C.

  When the page reading / displaying process is executed, the recommendation request unit 211 transmits the operation sequence operated so far, all the actions that can be operated next, and the user name (step S32). That is, the recommendation request unit 211 transmits evaluation requests for all the next operable actions to the recommendation transmission server 4. Note that all actions that can be operated next are read from the action DB 22 generated by the page reading / displaying process.

Thereafter, the recommendation display unit 212 receives N _OK (i, a) and N _NG (i, a), which are evaluation values for all actions, from the recommendation transmission server 4 (step S33). Note that i is a node corresponding to the current page. Let a be an action that can be operated next from the current page.

Then, the recommendation display unit 212 determines whether or not there is an action (a) such that “N _OK (i, a) + N _NG (i, a)> 0” among all the actions (step) S34). When it is determined that there is no action (a) that satisfies “N _OK (i, a) + N _NG (i, a)> 0” (step S34; No), the recommendation display unit 212 performs the next operation. In order to perform processing, the process proceeds to step S31.

On the other hand, when it is determined that there is an action (a) that satisfies “N _OK (i, a) + N _NG (i, a)> 0” (step S34; Yes), the recommendation display unit 212 displays the corresponding action. An evaluation value for is added to the current page and displayed (step S35). For example, the recommendation display unit 212 adds N _OK (i, a) and N _NG (i, a), which are evaluation values for the corresponding action, to the current page in association with the action. Then, the recommendation display unit 212 redisplays (refreshes) the added current page.

  And the recommendation display part 212 transfers to step S31 in order to process the next operation.

  As a result, the recommendation display unit 212 can display the number of persons as an evaluation value for all the actions that can be operated next to the current page, so that a recommendation for each action can be presented.

  FIG. 13B is a diagram illustrating a flowchart of a recommendation display process on the PC side that displays a recommendation with a probability. Note that the same operations as those in the recommendation display process illustrated in FIG. 13A are denoted by the same reference numerals, and the description of the overlapping operations is simplified.

  The recommendation request unit 211 executes a page reading / display process for a URL (step S31). The page reading / displaying process will be described later.

  When the page reading / displaying process is executed, the recommendation request unit 211 transmits the operation sequence operated so far, all the actions that can be operated next, and the user name (step S32).

  Thereafter, the recommendation display unit 212 receives P (a), which is an evaluation value for all actions, from the recommendation transmission server 4 (step S33A). In addition, P (a) is calculated by the formula (2) or the formula (3). Let a be an action that can be operated next from the current page.

  Then, the recommendation display unit 212 determines whether or not there is an action (a) that satisfies “P (a) ≠ undefined” among all actions (step S34A). If it is determined that there is no action (a) that satisfies “P (a) ≠ undefined” (step S34A; No), the recommendation display unit 212 proceeds to step S31 to process the next operation. .

  On the other hand, when it is determined that there is an action (a) that satisfies “P (a) ≠ undefined” (step S34A; Yes), the recommendation display unit 212 adds an evaluation value for the corresponding action to the current page. Are displayed (step S35A). For example, the recommendation display unit 212 adds P (a), which is an evaluation value for the corresponding action, to the current page in association with the action. Then, the recommendation display unit 212 redisplays (refreshes) the added current page.

  And the recommendation display part 212 transfers to step S31 in order to process the next operation.

  Thereby, since the recommendation display part 212 can display a probability as an evaluation value about all the actions which can be operated next on the present page, it can present the recommendation for every action.

  FIG. 13C is a flowchart of the page reading / display processing in FIGS. 13A and 13B.

  The recommendation request unit 211 determines whether a page is displayed (step S41). As an example, whether or not a page is displayed has an event attribute of “onVisit” in JavaScript. If it is determined that the page is not displayed (step S41; No), the recommendation request unit 211 repeats the determination process until the page is displayed.

  On the other hand, when it determines with the page being displayed (step S41; Yes), the recommendation request | requirement part 211 acquires all the links and all the forms from the displayed page (step S42). That is, the recommendation request unit 211 acquires all operation elements from the displayed page. For example, the value indicated by “length” is the number of links, and each link is a document. link [n] (n = 1, 2,..., length). The recommendation requesting unit 211 names links 1, 2,... The value indicated by “length” is the number of forms, and each form has a document. It is obtained by forms [n] (n = 1, 2,..., length). The recommendation request unit 211 names forms 1, forms 2,...

  And the recommendation request part 211 produces | generates action DB22 with respect to the acquired link and form (step S43). For example, the recommendation request unit 211 records the URL of the read page in association with all the acquired links and forms in the action DB 22. Then, the recommendation request unit 211 ends the page reading / displaying process.

[Procedure for operation recording processing on the recommendation recording server]
Next, the procedure of the operation recording process on the recommendation recording server 3 side will be described with reference to FIG. FIG. 14 is a flowchart of the operation recording process on the recommendation recording server side.

  First, the operation recording unit 311 determines whether an operation sequence, an evaluation, and a user name are received from the PC 1 (Step S51). When it is determined that the operation sequence, evaluation, and user name have not been received (step S51; No), the operation recording unit 311 repeats the determination process until the information is received.

  On the other hand, when it determines with having received the operation sequence, evaluation, and a user name (step S51; Yes), the operation recording part 311 compares the received operation sequence with recommendation DB32. Then, if there is a new transition destination (path) or a new URL (node) in the operation sequence, the operation recording unit 311 adds it to the recommendation DB 32 (step S52).

  Then, the operation recording unit 311 adds the received user name to the recommendation DB 32 for all nodes (URLs) that match the received operation sequence (step S53). The operation recording unit 311 adds the received user name to a location corresponding to the received evaluation. For example, if the operation sequence is top. html (form 1) → idea. html (form 2) → inventor. It is assumed that the user name is “A” and the evaluation is “OK”. In this case, the operation recording unit 311 adds the user name “A” to the location of OK 32 c for “top.html”, “idea.html”, and “inventor.html” as the URL 32 b.

  Then, the operation recording unit 311 ends the recommendation recording process.

[Procedure for recommendation calculation on the recommendation sending server]
Next, with reference to FIG. 15, the procedure of the recommendation calculation process in the recommendation transmission server 4 side is demonstrated. FIG. 15 is a diagram illustrating a flowchart of a recommendation calculation process on the recommendation transmission server side.

  First, the recommendation calculation unit 411 determines whether or not an operation sequence, all actions that can be operated next, and a user name have been received from the PC 2 (step S61). When it is determined that the operation sequence, all the actions that can be operated next and the user name have not been received (step S61; No), the recommendation calculation unit 411 repeats the determination process until the information is received.

  On the other hand, when it is determined that the operation sequence, all the actions that can be operated next and the user name have been received (step S61; Yes), the recommendation calculation unit 411 includes a path that matches the received operation sequence in the transition graph. It is determined whether or not there is (step S62). When it is determined that there is no matching path (step S62; No), the recommendation calculation unit 411 proceeds to step S61 to wait for the next reception.

  On the other hand, when it is determined that there is a matching path (step S62; Yes), the recommendation calculation unit 411 determines whether or not the following processing has been executed for all the next operable actions (step S63). ). When it is determined that the following processing has not been executed for all actions (step S63; No), the recommendation calculation unit 411 selects an action that has not been executed from all the actions that can be operated next ( Step S64). Then, the recommendation calculation unit 411 determines whether there is an action that matches the selected action in the node to be evaluated in the transition graph (step S65).

  If it is determined that there is no action that matches the selected action (step S65; No), the recommendation calculation unit 411 proceeds to step S63 to select the next action. On the other hand, if it is determined that there is an action that matches the selected action (step S65; Yes), the recommendation calculation unit 411 recommends a recommendation DB (normalized) 42 obtained by normalizing the recommendation DB 32 for the transition destination node of the selected action. Is acquired (step S66).

Then, the recommendation calculation unit 411 calculates P (a) as an evaluation value based on the recommendation DB (normalization) 42 and the social graph DB 43. Then, the recommendation calculation unit 411 transmits the calculated evaluation value to the transmission source PC 2 (step S67). Here, a indicates the selected action. For example, the recommendation calculation unit 411 calculates P (a) by the formula (2) or the formula (3). Note that the recommendation calculation unit 411 may calculate N _OK (i, a) and N _NG (i, a) as evaluation values based on the recommendation DB (normalization) 42 and the social graph DB 43. Here, i indicates a node corresponding to the page to be evaluated. a refers to the selected action. For example, the recommendation calculation unit 411 calculates N _OK (i, a) and N _NG (i, a) according to Expression (1), respectively.

  Subsequently, the recommendation calculation unit 411 proceeds to step S63 in order to select the next action. In step S63, when it is determined that the following processing has been executed for all actions (step S63; Yes), the recommendation calculation unit 411 ends the recommendation calculation processing.

  Thereby, the recommendation calculation part 411 can show the recommendation which considered the relationship with another user about the action which can be operated next to the evaluation object page on the PC 2.

[Example of screen]
Next, a screen of “patent application system” used as a web application in the embodiment will be described with reference to FIGS. 16A to 16F. 16A to 16F are diagrams illustrating examples of screens.

  The screen shown in FIG. 16A is a screen of the top URL “top.html” of “patent application system”. In “top.html”, an idea registration button, an inventor button, a strategy sheet button, a known example search result button, a temporary save button, a known example search request button, an application request button, and an end button are set. Each button has an element for each operation. For example, when the idea registration button is pressed (operated), the URL “idea.html” is displayed as the transition destination screen. When the inventor button is pressed (operated), the URL “inventor.html” is displayed as the transition destination screen. When the temporary save button is pressed (operated), the URL “save.html” is displayed as the transition destination screen. When the known example survey request button is pressed (operated), the URL “complete.html” is displayed as the transition destination screen.

  In “top.html”, a completion button, a good example button, and a bad example button are set. The “good example button” is pressed when the user thinks that a series of operations up to this screen has been executed correctly. Thereby, the operation sequence recording unit 111 records “OK” in the evaluation 12d corresponding to the corresponding URL 12b in the operation sequence DB 12. The “bad example button” is pressed when the user thinks that a series of operations up to this screen has not been executed correctly. Accordingly, the operation sequence recording unit 111 records “NG” in the evaluation 12d corresponding to the corresponding URL 12b in the operation sequence DB 12. The “complete button” is pressed when the user ends a series of operations. As a result, the operation record request unit 112 determines that a series of operations has been completed, and transmits an operation sequence or the like to the recommendation recording server 3. The completion button, good example button, and bad example button may be displayed or may not be displayed for each screen.

  The screen illustrated in FIG. 16B is a screen of URL “idea.html”. For example, “idea.html” is a transition made when the idea registration button is pressed in “top.html”. In “idea.html”, the same button as the URL “top.html” is set. Since the transition destination when each button is pressed is as described with reference to FIG. 16A, description thereof is omitted. In “idea.html”, a completion button, a good example button, and a bad example button are set. The functions of the completion button, good example button, and bad example button are as described with reference to FIG.

  The screen illustrated in FIG. 16C is a screen of URL “inventor.html”. “Inventor.html” is, for example, a transition made by pressing an inventor button in “idea.html”. The same button as the URL “top.html” is set in “inventor.html”. Since the transition destination when each button is pressed is as described with reference to FIG. 16A, description thereof is omitted. In “inventor.html”, a completion button, a good example button, and a bad example button are set. The functions of the completion button, good example button, and bad example button are as described with reference to FIG.

  The screen illustrated in FIG. 16D is a screen of URL “save.html”. “Save.html” is, for example, a transition made by pressing the temporary save button in “inventor.html”. The same button as the URL “top.html” is set in “save.html”. Since the transition destination when each button is pressed is as described with reference to FIG. 16A, description thereof is omitted. In “save.html”, a completion button, a good example button, and a bad example button are set. The functions of the completion button, good example button, and bad example button are as described with reference to FIG.

  The screen shown in FIG. 16E is a screen of URL “complete.html”. For example, “complete.html” is a transition made by pressing a known example survey request button in “search.html” (not shown). The same button as URL “top.html” is set in “complete.html”. Since the transition destination when each button is pressed is as described with reference to FIG. 16A, description thereof is omitted. In “complete.html”, a completion button, a good example button, and a bad example button are set. The functions of the completion button, good example button, and bad example button are as described with reference to FIG.

  The screen illustrated in FIG. 16F is a screen of URL “error.html”. For example, in the “patent application system”, it is assumed that temporary storage is performed before a known example search request or application request is made. If the well-known example survey request button is pressed before temporary storage, “error.html” is displayed. The same button as the URL “top.html” is set in “error.html”. Since the transition destination when each button is pressed is as described with reference to FIG. 16A, description thereof is omitted. In “error.html”, a completion button, a good example button, and a bad example button are set. The functions of the completion button, good example button, and bad example button are as described with reference to FIG.

[Example of recommendation display]
Next, an example of recommendation display by the recommendation display unit 212 will be described with reference to FIGS. 17A and 17B. 17A and 17B are diagrams illustrating an example of recommendation display. FIG. 17A is a recommendation display when the recommendation is displayed by the number of persons, and FIG. 17B is a recommendation display when the recommendation is displayed by the probability. The URL “top.html” has a plurality of buttons that can be operated next.

  As illustrated in FIG. 17A, the recommendation display unit 212 displays the number of people as an evaluation value for each button corresponding to an action. Here, for the idea registration button, the recommendation display unit 212 displays “One of the people who clicked here is good, 0.8 is bad”. For the inventor button, the recommendation display unit 212 displays “one of the people who click here is good and one is bad”. That is, the recommendation of the idea registration button and the inventor button is displayed by the number of persons.

  As illustrated in FIG. 17B, the recommendation display unit 212 displays the probability as the evaluation value for each button corresponding to the action. Here, for the idea registration button, the recommendation display unit 212 displays “Click here for OK 55% (next is idea.html)”. For the inventor's button, the recommendation display unit 212 displays “Click here, OK 50% (next is inventor.html)”. That is, the recommendation of the idea registration button and the inventor button is displayed with probability.

[Effect of Example]
According to the above embodiment, the recommendation transmission server 4 receives an evaluation request for an operation candidate that can be operated next for a predetermined node in the operation sequence being operated. And the recommendation transmission server 4 calculates the evaluation value of the operation candidate which can be operated next about the node which received the operation evaluation request | requirement based on recommendation DB42 and social graph DB43, respectively. The recommendation DB 42 indicates the evaluation of each user for each operation on each node in the operation sequence. Social graph DB43 shows the distance between each user. The recommendation transmission server 4 associates each calculated evaluation value with each operation candidate and transmits it to the transmission source. According to this configuration, the recommendation transmission server 4 calculates the evaluation value of the operation candidate that can be operated next in consideration of the relationship between the user who is operating and each user, and transmits the node for the operation sequence being operated. Since it is transmitted to the original, it becomes possible to present a recommendation for each operation candidate. That is, the recommendation transmission server 4 can recommend a series of operations of the user who is operating.

  Moreover, according to the said Example, the recommendation transmission server 4 receives the operation sequence operated by the user in operation, and the several operation candidate which can be operated next. And the recommendation transmission server 4 reads the evaluation in each user for every operation candidate which can be operated next to the received operation sequence based on the recommendation DB. Then, the recommendation transmission server 4 calculates an evaluation value based on the read evaluation for each user and information indicating the distance between each user of the social graph DB 43 and the user in operation. According to such a configuration, the recommendation transmission server 4 calculates the evaluation value of the operation candidate that can be operated next in consideration of the distance between the user who is operating and each user, and thus has a strong connection with the user who is operating. An evaluation value indicating how many users are included can be calculated. As a result, the recommendation transmission server 4 can present a recommendation that increases the reliability of a series of operations to the operating user.

  Further, according to the above embodiment, the recommendation DB 42 stores information indicating an evaluation of “OK” or an evaluation of “NG” as an evaluation for each user for each operation on each node of the operation sequence. The recommendation transmission server 4 calculates an evaluation value that is evaluated as “OK” and an evaluation value that is evaluated as “NG”, for each operation candidate that can be operated next to the received operation sequence. Then, the recommendation transmission server 4 calculates the probability of “OK” evaluation using the evaluation value for “OK” evaluation and the evaluation value for “NG” evaluation. According to such a configuration, the recommendation transmission server 4 uses the probability of “OK” evaluation obtained from the evaluation of “OK” and the evaluation of “NG” for each user as the evaluation value of the operation candidate that can be operated next. It can be calculated. As a result, the recommendation transmission server 4 can present to the user in operation a recommendation that increases the reliability of the majority of the series of operations.

  Further, according to the above embodiment, the recommendation transmission server 4 evaluates “OK” and evaluation values “NG” so as to highly evaluate either the subsequent operation or the latest operation. After setting the weight to, the probability of evaluating “OK” is calculated. With this configuration, the recommendation transmission server 4 can present a recommendation according to the nature of a series of operations.

  Further, according to the above-described embodiment, the recommendation transmission server 4 transmits the probability of evaluation of “OK” displayed in correspondence with the next operable operation candidate. According to this configuration, the transmission destination can display the probability of “OK” evaluation in correspondence with the next operable operation candidate, so that the next operation can be efficiently selected for the user who is operating.

[Others]
In the above embodiment, the operation support system 9 has been described as having a configuration in which the recommendation recording server 3 and the recommendation transmission server 4 are separated. However, the operation support system 9 is not limited to this, and may be configured to integrate two servers into one server.

  In the above embodiment, the PC 1 is described as a device that requests operation recording of the operation sequence of each user, and the PC 2 is described as a device that displays a recommendation of another user's operation sequence. However, the operation support system 9 is not limited to this, and the PC 1 and the PC 2 may include both functions.

  Further, in the above embodiment, the operation support system 9 executes the operation recording process in the recommendation recording server 3 and then executes the recommendation calculation process in the recommendation transmission server 4. However, the operation support system 9 is not limited to this, and may be executed simultaneously (in parallel). According to such a configuration, the operation support system 9 can efficiently perform a series of operation recommendations.

  Further, the recommendation recording server 3 can be realized by mounting the above-described functions such as the control unit 31 and the recommendation DB 32 on an information processing apparatus such as a known personal computer or workstation. Further, the recommendation transmission server 4 can be realized by mounting the above-described functions such as the control unit 41, the recommendation DB 42, and the social graph DB 43 in an information processing apparatus such as a known personal computer or workstation.

  In addition, each component of the illustrated apparatus does not necessarily need to be physically configured as illustrated. In other words, the specific mode of device distribution / integration is not limited to that shown in the figure, and all or part of the device is functionally / physically distributed / integrated in arbitrary units according to various loads and usage conditions. Can be configured. For example, the operation sequence recording unit 111 and the operation record requesting unit 112 may be integrated as one unit. Further, the recommendation request unit 211 and the recommendation display unit 212 may be integrated as one unit. On the other hand, the recommendation calculation unit 411 is distributed to a first calculation unit that calculates a good example evaluation value and a bad example evaluation value, and a second calculation unit that calculates a probability based on the recommendation evaluation formula. You may do it. The recommendation DB 32 may be connected as an external device of the recommendation recording server 3 via a network. Moreover, you may make it connect recommendation DB42 and social graph DB43 as an external apparatus of the recommendation transmission server 4 via a network.

  The various processes described in the above embodiments can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. Therefore, an example of a computer that executes an operation support program that realizes the same function as the recommendation transmission server 4 illustrated in FIG. 1 will be described below. FIG. 18 is a diagram illustrating an example of a computer that executes an operation support program.

  As illustrated in FIG. 18, the computer 200 includes a CPU 203 that executes various arithmetic processes, an input device 215 that receives input of data from the user, and a display control unit 207 that controls the display device 209. The computer 200 also includes a drive device 213 that reads a program or the like from a storage medium, and a communication control unit 217 that exchanges data with another computer via a network. The computer 200 also includes a memory 201 that temporarily stores various types of information and an HDD 205. The memory 201, CPU 203, HDD 205, display control unit 207, drive device 213, input device 215, and communication control unit 217 are connected by a bus 219.

  The drive device 213 is a device for the removable disk 211, for example. The HDD 205 stores an operation support program 205a and operation support related information 205b.

  The CPU 203 reads the operation support program 205a, expands it in the memory 201, and executes it as a process. Such a process corresponds to each functional unit of the recommendation transmission server 4. The operation support related information 205 b corresponds to the recommendation DB (normalization) 42 and the social graph DB 43. For example, the removable disk 211 stores each piece of information such as the operation support program 205a.

  Note that the operation support program 205a is not necessarily stored in the HDD 205 from the beginning. For example, the program is stored in a “portable physical medium” such as a flexible disk (FD), a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card inserted into the computer 200. Then, the computer 200 may read out and execute the operation support program 205a from these.

  The following supplementary notes are further disclosed regarding the embodiment described above and its modifications.

(Supplementary note 1)
Receiving an evaluation request for an operation candidate that can be operated next for a predetermined node in the operation sequence being operated,
Based on the first information indicating the evaluation of each user for each operation for each node in the operation sequence and the second information indicating the distance between the users, the node that has received the operation evaluation request can be operated next. Calculate the evaluation value of each operation candidate,
Each evaluation value calculated by the calculation process is associated with each operation candidate and transmitted to the transmission source.
An operation support program characterized by causing each process to be executed.

(Supplementary Note 2) The receiving process receives an operation sequence operated by another user and a plurality of operation candidates that can be operated next,
Based on the first information, the calculation process reads the evaluation for each user for each operation candidate that can be operated next to the received operation sequence, and the evaluation for each read user and the second The operation support program according to appendix 1, wherein an evaluation value is calculated based on information indicating a distance between each user of the information and the different user.

(Supplementary Note 3) The first information stores information indicating the first evaluation or the second evaluation as an evaluation in each user for each operation on each node of the operation sequence,
The calculation processing calculates an evaluation value to be the first evaluation and an evaluation value to be the second evaluation for each operation candidate that can be operated next to the received operation sequence, and The operation support program according to supplementary note 2, wherein the probability of the first evaluation is calculated using the evaluation value to be evaluated and the evaluation value to be the second evaluation.

(Supplementary Note 4) In the calculation process, a weight is set to the evaluation value to be the first evaluation and the evaluation value to be the second evaluation so as to highly evaluate either the subsequent operation or the most recent operation. In addition, the operation support program according to appendix 1, wherein a probability of the first evaluation is calculated.

(Supplementary Note 5) The operation according to Supplementary Note 3 or Supplementary Note 4, wherein the transmission process transmits the probability of the first evaluation displayed in correspondence with the next operable operation candidate. Support program.

(Appendix 6)
Receiving an evaluation request for an operation candidate that can be operated next for a predetermined node in the operation sequence being operated,
Based on the first information indicating the evaluation of each user for each operation for each node in the operation sequence and the second information indicating the distance between the users, the node that has received the operation evaluation request can be operated next. Calculate the evaluation value of each operation candidate,
Each evaluation value calculated by the calculation process is associated with each operation candidate and transmitted to the transmission source.
An operation support method characterized by executing each process.

(Supplementary Note 7) A receiving unit that receives an evaluation request for an operation candidate that can be operated next with respect to a predetermined node of the operation sequence being operated
Based on the first information indicating the evaluation of each user for each operation for each node in the operation sequence and the second information indicating the distance between the users, the node that has received the operation evaluation request can be operated next. A calculation unit for calculating an evaluation value of each operation candidate;
A transmission unit that transmits each evaluation value calculated by the calculation processing to each transmission candidate in association with each operation candidate;
An information processing apparatus comprising:

1, 2 PC
3 Recommendation Recording Server 4 Recommendation Transmission Server 5 Web Server 6 Internet 9 Operation Support System 11 Browser 12 Operation Sequence DB
111 Operation Sequence Recording Unit 112 Operation Record Request Unit 21 Browser 22 Action DB
211 recommendation request unit 212 recommendation display unit 31, 41 control unit 32 recommendation DB
311 Operation Recording Unit 42 Recommendation DB (Normalized)
43 Social Graph DB

Claims (6)

On the computer,
Receiving an evaluation request for an operation candidate that can be operated next for a predetermined node in the operation sequence being operated,
Based on the first information indicating the evaluation of each user for each operation for each node in the operation sequence and the second information indicating the distance between the users, the node that has received the operation evaluation request can be operated next. Calculate the evaluation value of each operation candidate,
Each evaluation value calculated by the calculation process is associated with each operation candidate and transmitted to the transmission source.
An operation support program characterized by causing each process to be executed. The receiving process receives an operation sequence operated by another user and a plurality of operation candidates that can be operated next,
Based on the first information, the calculation process reads the evaluation for each user for each operation candidate that can be operated next to the received operation sequence, and the evaluation for each read user and the second The operation support program according to claim 1, wherein an evaluation value is calculated based on information indicating a distance between each user of the information and the different user. The first information stores information indicating a first evaluation or a second evaluation as an evaluation in each user for each operation on each node of the operation sequence;
The calculation processing calculates an evaluation value to be the first evaluation and an evaluation value to be the second evaluation for each operation candidate that can be operated next to the received operation sequence, and 3. The operation support program according to claim 2, wherein a probability of the first evaluation is calculated using an evaluation value to be evaluated and an evaluation value to be the second evaluation. In the process of calculating, after setting a weight to the evaluation value to be the first evaluation and the evaluation value to be the second evaluation so as to highly evaluate either the subsequent operation or the most recent operation, The operation support program according to claim 3 , wherein a probability of the first evaluation is calculated. Computer
Receiving an evaluation request for an operation candidate that can be operated next for a predetermined node in the operation sequence being operated,
Based on the first information indicating the evaluation of each user for each operation for each node in the operation sequence and the second information indicating the distance between the users, the node that has received the operation evaluation request can be operated next. Calculate the evaluation value of each operation candidate,
Each evaluation value calculated by the calculation process is associated with each operation candidate and transmitted to the transmission source.
An operation support method characterized by executing each process. A receiving unit that receives an evaluation request for an operation candidate that can be operated next with respect to a predetermined node of the operation sequence being operated;
Based on the first information indicating the evaluation of each user for each operation for each node in the operation sequence and the second information indicating the distance between the users, the node that has received the operation evaluation request can be operated next. A calculation unit for calculating an evaluation value of each operation candidate;
A transmission unit that transmits each evaluation value calculated by the calculation processing to each transmission candidate in association with each operation candidate;
An information processing apparatus comprising:

Images