User selection device

Abstract

To provide a user selection device capable of selecting another user who is expected to provide an appropriate answer to a question from a user.SOLUTION: In a user selection device 100, a similarity calculation unit 103 calculates the similarity between a post Q (post content) posted by a user and user information of another user (such as a profile and post content of the other user), a relationship calculation unit 105 calculates the relationship between the user and the other user based on the similarity, and a selection unit 107 selects the other user as a candidate user on the basis of the calculated relationship and similarity. This configuration enables selection of the candidate user on the basis of the relationship between the users in addition to the similarity of the post content. The candidate user selected with consideration of the relationship is expected to provide an answer to the post for the one user.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a user selection device for selecting a user. [Background technology]

[0002] Patent document 1 describes that based on the acquired question information and behavioral information related to behavioral history on the network, users who use the service are selected as potential answerers who will provide answers to the questions. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-200713 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, even if a user who is not acquainted with the questioner becomes a candidate answerer, there is a possibility that an appropriate answer will not be obtained.

[0005] Therefore, an object of the present invention is to provide a user selection device capable of selecting an appropriate other user. [Means for solving the problem]

[0006] The user selection device of the present invention includes a similarity calculation unit that calculates the similarity between the content posted by a user and the user information of other users, a relationship calculation unit that calculates the relationship between the user and other users, and a selection unit that selects the other users as candidate users based on the relationship and the similarity. Effect of the Invention

[0007] According to the present invention, an appropriate user can be selected. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing an overview of the problem to be solved by the present disclosure. [Diagram 2] FIG. 1 is a schematic diagram illustrating making a user selection by a user selection device 100 of the present disclosure. [Diagram 3] 1 is a diagram showing a system configuration of a Q&A posting service system including a user selection device 100 according to the present disclosure. [Figure 4] 1 is a block diagram showing a functional configuration of a user selection device 100 according to the present disclosure. [Diagram 5] 1 is a schematic diagram showing an overview of the processing of a user selection device 100. FIG. [Figure 6] FIG. 13 is a diagram showing an overview of generation of a user's feature vector. [Figure 7] FIG. 13 is a diagram showing the second half of the process outline of the user selection device 100. [Figure 8] 13 is a flowchart showing the operation of a relationship calculation unit 105. [Figure 9] FIG. 11 is a diagram illustrating an example of input data. [Figure 10] FIG. 13 is a diagram showing a shaping process for nomination information. [Figure 11] FIG. 13 is a diagram showing the formatting process for nomination information, Q&A exchange history, and profile browsing history. [Figure 12] FIG. 13 is a diagram illustrating a scaling process. [Figure 13] FIG. 13 is a diagram illustrating a linear combination process. [Figure 14] FIG. 13 illustrates a scaling process for data after linear combination. [Figure 15] FIG. 13 is a conceptual diagram showing a linear combination process when nomination information is further added. [Figure 16] 13 is a flowchart showing a process of correcting a similarity based on a relationship. [Figure 17] FIG. 13 is a diagram illustrating an example of correcting the similarity using the relationship between users. [Figure 18] FIG. 13 is a schematic diagram showing correction by multiplication of interest level and similarity level. [Figure 19] FIG. 13 is a schematic diagram showing a process in which the interest level and the similarity level are each given a boost. [Figure 20] 1 is a diagram illustrating an example of a hardware configuration of a user selection device 100 according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] The embodiments of the present disclosure will be described with reference to the accompanying drawings. Whenever possible, the same parts are designated by the same reference numerals and redundant description will be omitted.

[0010] FIG. 1 is a diagram illustrating an overview of the problem to be solved by the present disclosure. In the diagram, user A posts a post Q to a QA posting service server. In the conventional technology, the QA posting service server selects users C and D with high similarity as candidates and recommends users C and D to user A. The QA posting service server then recommends (sends) post Q posted by user A to users C and D.

[0011] It is expected that users C and D will reply to post Q, but depending on the relationship between user A and users C and D, there may be a situation where a reply from users C and D is not obtained.

[0012] As mentioned above, there is a psychological hurdle for other users, such that they are reluctant to reply to posts Q from users they do not know or with whom they are not on good terms.

[0013] 2 is a schematic diagram showing user selection by the user selection device 100 of the present disclosure. As shown in the figure, the user selection device 100 corrects the similarity calculated from the content of the post Q posted by the user and the user information of other users based on the relationship between the users, and then selects a candidate user. The user selection device 100 of the present disclosure does not select a user by using the calculated similarity as it is, but corrects the similarity based on a score representing the relationship between the users, and then selects a user.

[0014] FIG. 2(a) is a diagram showing relationships between users. The user selection device 100 selects several users that are deemed appropriate from users A to E based on the relationships. In FIG. 2(b), the user selection device 100 corrects the similarity of each user based on the relationships. Then, the user selection device 100 selects the user with the highest corrected similarity. In the present disclosure, the user selection device 100 selects user B, whose similarity of 0.6 is corrected to 0.8.

[0015] 3 is a diagram showing a system configuration of a QA posting service system including a user selection device 100 of the present disclosure. As shown in the figure, this QA posting service system includes the user selection device 100 and a QA posting service server 200. The QA posting service server 200 is configured to be communicatively connected to the user selection device 100 and a user terminal 300. This QA posting service server 200 can accept a posted Q from the user terminal 300 of one user and can accept answers to the posted Q from the user terminals 300 of other users.

[0016] When a post Q of a user is accepted by the QA posting service server 200, the user selection device 100 selects another user who is similar to the selected user and whose relationship with the selected user satisfies a predetermined condition, and transmits a notification to the user terminal 300 of the other user to prompt the user to reply to the post Q. In the present disclosure, the user selection device 100 and the QA posting service server 200 are represented as separate devices, but this is not limiting, and the QA posting service server 200 may include the functions of the user selection device 100.

[0017] 4 is a block diagram showing a functional configuration of a user selection device 100 according to the present disclosure. As shown in the figure, the user selection device 100 includes a posted information characteristic vectorization unit 101, a user information characteristic vectorization unit 102, a similarity calculation unit 103, a storage unit 104, a relationship calculation unit 105, a correction unit 106, and a selection unit 107.

[0018] The posted information feature vectorization unit 101 is a part that converts the content of a post Q posted by a user into a feature vector using a language model. The content of a post Q may include a title, a main text, tags, text in an attached file, and the like. The language model is a publicly known one, and although Universal Sentence Encoder is used in this disclosure, feature vectorization may be performed using other publicly known language models such as Doc2vec or Sentence Bert. Of course, the language model is not limited to these.

[0019] The user information feature vectorization unit 102 is a part that generates a feature vector based on the user information of other users. In the present disclosure, the user information is the posting contents registered by other users and the profile information of other users. The profile information includes the user's affiliation, job title, special skills, past efforts, awards, qualifications, training history, hard skills, soft skills, the work content that the user wants to do, work that matches the hard skills, soft skills, and the work content that the user wants to do, a self-introduction, text in the self-introduction attachment, past QA posting contents, the number of "Like" stamps obtained for past posting contents, and the like. Note that weighting processing may be performed so that the posting contents that have obtained the "Like" stamp are strongly reflected in the user feature vector. Any other text information generated by other users may be used. Weighting processing may also be performed based on the posting date.

[0020] The similarity calculation unit 103 is a part that calculates the similarity between the feature vector of the posted information posted by the user and the feature vector of the posted information and / or profile information posted by other users. The similarity is, for example, cosine similarity, but is not limited thereto.

[0021] The storage unit 104 is a part that stores interactions between users in the Q&A posting service (e.g., Q&A exchanges, profile page browsing history), and designation information that is connection information input by users. That is, the storage unit 104 stores each user's profile (including designation information indicating connections), Q&A exchange history (who responded to which user's post), and profile browsing history (who viewed which user's profile).

[0022] The relationship calculation unit 105 is a part that calculates relationship information indicating a relationship between one user and another user based on the interaction and connection information stored in the storage unit 104. For example, this relationship information is expressed by a weighted directed graph, but is of course not limited to this.

[0023] The correction unit 106 is a part that corrects the similarity calculated by the similarity calculation unit 103 based on the relationship information calculated by the relationship calculation unit 105.

[0024] The selection unit 107 is a part that selects one or more other users based on the corrected similarity.

[0025] The notification unit 108 transmits a notification to the user terminal 300 of the selected one or more other users to prompt them to reply to the post Q. The notification unit 108 sends a notification to prompt them to reply to the post Q, but may also send the notification before the post Q is posted if the post Q is in a draft state. This allows the candidate user to be notified before a reply from a user other than the candidate user, and to expect the reply.

[0026] Next, an overview of the processing of the user selection device 100 configured as above will be described. Fig. 5 is a schematic diagram showing the processing overview. Fig. 5(a) is a diagram showing how the posted information feature vectorization unit 101 converts the content of a post Q posted by a user into a feature vector using a language model.

[0027] 5(b) is a diagram showing how the user information feature vectorization unit 102 converts user information into a feature vector by a language model. In the figure, the user information feature vectorization unit 102 applies a self-introduction, a work history (content expressed in text), and posted information posted by the user or answer information to posted information of other users to a language model (the same as the language model of the posted information feature vectorization unit 101) to calculate a feature vector.

[0028] Then, the user information characteristic vectorization unit 102 weights the characteristic vectors obtained from each of them to obtain a characteristic vector for the user.

[0029] Fig. 6 is a diagram showing an overview of the generation of a user's feature vector. In Fig. 6, a feature vector is obtained using a self-introduction text and two posted contents as examples. As shown in Fig. 6, the user information feature vectorization unit 102 performs weighted averaging processing on the feature vector obtained from the self-introduction text and the two feature vectors obtained from the posted contents to generate a feature vector for the user.

[0030] When performing weighted averaging, the user information feature vectorization unit 102 sets the weight so that the most recent posting content is given a larger weight. For example, in FIG. 6, the recently updated self-introduction text and the posting content whose date is close to the current time (the current date and time is 2023 / 04 / 20), such as 2023 / 04 / 01, are given larger weights of 0.5 and 0.4. On the other hand, the user information feature vectorization unit 102 gives a smaller value of 0.1 to the posting content whose date is slightly older, such as 2023 / 03 / 16. By doing so, the user information feature vectorization unit 102 creates a feature vector based on user information while suppressing changes in the user's tendency over time (taking into account the recent user's trend).

[0031] 5(c) is a diagram showing a similarity calculation process by the similarity calculation unit 103. As shown in the figure, the similarity calculation unit 103 calculates a similarity of 0.6 between the feature vector of post Q by user A and the feature vector of user B. Similarly, the similarity calculation unit 103 calculates a similarity of 0.3 between the feature vector of post Q and the feature vector of user C.

[0032] Fig. 7 is a diagram showing the latter half of the processing overview of the user selection device 100. Fig. 7(a) is a diagram showing the relationship between users calculated as a weighted directed graph by the relationship calculation unit 105. As shown in Fig. 7(a), the validity graph is based on past interactions and input connection information. The interactions are Q&A exchange history, profile browsing history, etc. Furthermore, the connection information is nomination information generated by the user's self-reporting. The nomination information will be described in detail later.

[0033] 7(b) is a diagram showing the similarity correction performed by the correction unit 106 using the inter-user relationship information. As shown in the figure, the correction unit 106 corrects, for example, the similarity between user A and user B, which is 0.6, to 0.8 based on the relationship information. Similarly, the similarity between user A and user C is corrected.

[0034] As shown in FIG. 7(b), user B having the higher corrected similarity is selected.

[0035] Next, the details of the relationship calculation process between users will be described with reference to a flowchart of the operation of relationship calculation unit 105.

[0036] The relationship calculation unit 105 refers to the storage unit 104, acquires nomination information of other users by one user, tallying up QA exchange history, and tallying up profile browsing history, and acquires them as input data (S101, S102, S103). The storage unit 104 stores each user's profile (including nomination information indicating a connection), QA exchange history (who responded to which user's post), and profile browsing history (who viewed which user's profile).

[0037] Examples of input data include the data shown in Fig. 9. Consider five users, A, B, C, D, and E, in Fig. 9. The relationship calculation unit 105 uses three pieces of data to calculate the relationship between the users.

[0038] -Users who have similar skills to you, as entered on your profile page (designated information) -Which user replied to whose Q posted in the Q&A interaction history (Q&A interaction history) - Which user has viewed which user's profile page (how many times and for how long) in the profile page browsing history (profile browsing history) In the present disclosure, the input data includes data indicating the skills possessed by a user similar to that of the user, as shown in FIG. 9(a), data indicating the Q&A exchange history, as shown in FIG. 9(b), and data indicating the profile viewing history, as shown in FIG. 9(c).

[0039] In Fig. 9(a), the nominee column and the nominee column are associated with each other. This indicates that user A nominates users C and D, and that they have similar skills to user A. This data is stored in user A's profile as nomination information, and the profile of each user is stored in storage unit 104.

[0040] In Fig. 9(b), the poster of A and the poster of Q are associated as the QA interaction history. It is shown that user C has replied to user D's post four times and to user E's post three times. It is also shown that user A has not replied to anyone's post. φ indicates no reply.

[0041] In Fig. 9(c), the viewer and the person who was viewed are associated as the profile viewing history. It shows that user A viewed user B's profile for 13 seconds, user C's for 21 seconds, and user D's for 14 seconds. It also shows that user D did not view anyone's profile. φ indicates no viewing.

[0042] Then, the relationship calculation unit 105 performs a formatting process on the acquired and compiled input data such as the profile, exchange history, and browsing history (S104).

[0043] The formatting process will now be described with reference to Fig. 10, which shows the formatting process for nomination information.

[0044] History information such as Q&A interaction history and profile viewing history exists as a pair of a user and a value, as shown below.

[0045] (u, n): Answered user u's Q n times (v, m): Viewed user v's profile for a total of m seconds On the other hand, for nomination information (nomination of possessed skills), although there is a user, there is no value. Therefore, in order to handle nomination information and history information in the same framework, the nomination information is assigned a value of 1 to indicate that a nomination has been made.

[0046] In other words, unlike the Q&A exchange history and profile viewing history, the nomination information does not have the concept of number of times. Therefore, the data structure of the nomination information is different from that of history information such as the Q&A exchange history. The formatting process is a process to align the data structure.

[0047] Fig. 10(a) is a diagram showing nomination information. Fig. 10(b) shows the nomination information with a value of 1. In this way, by associating a nominee with a value in the nominee field in the nomination information, the data structure can be made the same as other history information.

[0048] Furthermore, the relationship calculation unit 105 performs a formatting process shown in Fig. 11. Up until now, the symbol φ representing an empty set was used to indicate that no data exists (no one has been nominated / no one has responded to anyone's post / no one has viewed anyone's profile). If the value 0 is used to indicate no nomination / no response / no viewing, the above data can be expressed without using φ.

[0049] That is, in Fig. 11, for items for which no data exists, 0 is assigned. For example, in Fig. 11(a), user A, who is the nominator, has not nominated user B and user E. Therefore, the relationship calculation unit 105 generates data indicating that these users have not been nominated by assigning 0 to them.

[0050] In FIG. 11(b), the relationship calculation unit 105 assigns a value of 0 to users who have no interaction history. For example, no one has responded to a post by user A. Therefore, in the Q poster column, a value of 0 is assigned to all users. Similarly, a value of 0 is assigned to other users who have not responded.

[0051] 11C, the relationship calculation unit 105 assigns a value of 0 to users who have no profile browsing history. For example, the profile of user D has not been viewed by anyone, so a value of 0 is assigned to each user.

[0052] Next, the relationship calculation unit 105 performs a scaling process on each piece of data after performing the data shaping process (S105).

[0053] The data of the nomination information can only take on values ​​of 0 or 1, whereas the data of the history information can take on values ​​of 0 or N (N is a natural number). In other words, the scale of values ​​may differ greatly between the nomination information and the history information. In particular, the value of the browsing history information of the profile is likely to take on a particularly large value compared to other data. Since performing subsequent processing in this state will affect accuracy, the relationship calculation unit 105 performs a value scaling process on the data of the history information. In the example of FIG. 12, Min-Max standardization is applied as a method of scaling, but methods such as z-score standardization and robust z-score standardization may also be used.

[0054] FIG. 12(a) is a diagram showing scaling processing (Min-Max standardization) for QA interaction history information. As shown in the diagram, User C provided the most replies to User D's posts. This is set as the maximum value, and scaling processing is performed between 0 and 1. Scaling processing is performed by multiplying each number by the reciprocal of the maximum value. FIG. 12(b) is a diagram showing the QA interaction history information after this scaling processing. Here, User C's value for User D is 1. Meanwhile, User B's value for User A is 0.5.

[0055] Fig. 12(c) is a diagram showing a scaling process for a profile browsing history. In the figure, User C viewed User D's profile for the longest time, at 116 seconds. Therefore, the relationship calculation unit 105 performs a scaling process based on that, similar to Fig. 12(a).

[0056] Next, the relationship calculation unit 105 performs a linear combination process on each data (S106). That is, the relationship calculation unit 105 linearly combines each data (multiplying each data by a weight w and then adding them up) and integrates them into one data. This weight w is a parameter for determining which data is to be emphasized in calculating the relationship between users, and a value is set such that the sum of all parameters is 1. For example, in the example of FIG. 13, w1=0.1, w2=0.2, and w3=0.7, and the relationship between users is calculated by placing the most importance on the data of the profile browsing history. This parameter is set by the operator of the user selection device 100.

[0057] Figure 13(a) shows the nomination information, QA exchange history information, and profile browsing history information that indicate users who have similar skills to the user. The data in Figure 13(b) is obtained by linearly combining the weights w1 = 0.1, w2 = 0.2, and w3 = 0.7, respectively.

[0058] 13(b) is a diagram showing the linear combination processed data, which is composed of a reference source and a reference destination. The reference source indicates a user, and the reference destination indicates a user referred to by the reference source user, that is, indicates the degree of interest (degree of interest) of the reference source user in the reference destination user.

[0059] The relationship calculation unit 105 performs a scaling process on the data after the linear combination (S107). That is, the relationship calculation unit 105 applies Min-Max standardization to the obtained data so that the value range is between 0 and 1. In the example of FIG. 13, Min-Max standardization is used, and weights w are set so that the total value of all weights w is 1, so that the obtained data is already within the range of 0 to 1. However, if a scaling method other than Min-Max standardization is used in step S105, it is not guaranteed that the obtained data is within the range of 0 to 1, so a final scaling process is required. Therefore, if a scaling method according to Min-Max standardization is used, the process of step S107 is not essential.

[0060] Fig. 14(a) shows the linearly combined data. The relationship calculation unit 105 performs Min-Max standardization on this data. Fig. 14(b) shows the scaled data. As shown in the figure, the value of user C for user D is the maximum value, and since the value is 1, there is no change in the value itself.

[0061] Fig. 14(c) is a diagram expressed as a directed graph based on the data in Fig. 14(b). The direction of the arrow indicates the reference destination. In Fig. 14(c), for example, the degree of relationship between user A and user B is 0.078.

[0062] 13 shows the linear combination process based on the nomination information indicating users who have similar skills to the user, the QA exchange history information, and the profile browsing history information, but the process is not limited to this. For example, users who have similar values ​​or ways of thinking to the user may be stored in the user's profile information as nomination information, and the stored information may be used as another nomination information in the linear combination process.

[0063] FIG. 15 is a conceptual diagram showing the linear combination process when further nomination information is added. Here, the weight w 1-1 (Weight w1) has weight w 1-2 The data is added to the profile information, and linear combination processing is performed using the data. In addition, other users who are frequently contacted for work and other users who are frequently contacted outside of work may be stored in the profile information as nomination information, and these may be further added or replaced with each other to perform data shaping processing, scaling processing, and linear combination processing. Note that the numerical values ​​of the vectors in FIG. 15 are the same as those in FIG. 13 for the sake of convenience, but they are not related to those in FIG. 13.

[0064] Further, information indicating relationships may also include following / following back on social media or similar information, and by using these as input data, it is also possible to perform data formatting processing, etc. to generate information indicating relationships.

[0065] Next, a process of correcting the similarity based on the relationship obtained as described above will be described. FIG. 16 is a flowchart showing the correction process. The correction unit 106 adds a fixed value (additional value) to the relationship between users calculated by the relationship calculation unit 105 (S201). The correction unit 106 adds a fixed value (additional value) to the similarity calculated by the similarity calculation unit 103 (S202). The correction unit 106 multiplies the relationship between users by the similarity to correct the similarity (S203).

[0066] The selection unit 107 selects a user based on the multiplied value (corrected similarity) (S204).

[0067] In this way, the correction unit 106 corrects the similarity based on the relationship, and the selection unit 107 selects a user based on the corrected value, thereby making it possible to select a user while taking the relationship into consideration.

[0068] Fig. 17 is a diagram showing an example of correcting the similarity using the relationship between users. Here, five users, A, B, C, D, and E, are considered. It is assumed that the values ​​in the table shown in Fig. 17(a) are obtained as the relationships between these users.

[0069] Post Q posted by user A(Q A ) and other users (users B, C, D, and E) (sim(Q A Let us assume that when we calculate the user ID (expressed as user ID, user), we obtain the following value:

[0070] sim(Q A , B) =0.6 sim(Q A , C) =0.9 sim(Q A , D) =0.4 sim(Q A , E) =0.3 Fig. 17(b) is a schematic diagram of the similarity. As shown in Fig. 17(b), the similarity between the feature vector of post Q registered by user A and the feature vector of user B is 0.6. The same is true for other users C to E.

[0071] FIG. 18 is a schematic diagram showing correction by multiplying the interest level and the similarity level. FIG. 18(a) shows a directed graph showing the interest level of each user in user A among the relationships between users. This validity graph focuses on the reference destination of user A among the data showing the relationships between users, and is a directed graph showing the interest level of each user in user A among the directed graphs shown in FIG. 14(c). That is, it is a value (interest level) showing how much interest other users other than user A have in user A. It is considered that the larger the value, the more likely it is that a user will reply to the post Q created by the user. In the present disclosure, the relationship between user A and other users shown in FIG. 18(a) is the interest level of each user in user A.

[0072] Fig. 18(b) is a diagram showing the similarity between users. Fig. 18(c) is a diagram showing the similarity after correction by multiplying the similarity of the feature vectors of each user by the interest level. In this diagram, user B has an interest level of 0.275 for user A, and by multiplying this by the similarity level of 0.6, a corrected similarity level of 0.165 is obtained. The same applies to other users.

[0073] The selection unit 107 selects a user using this corrected similarity, thereby making it possible to select a user in consideration of the interest level (relationship).

[0074] Incidentally, as shown in Figure 18, it seems intuitive to correct the similarity by multiplying the interest level and similarity level, but if the multiplication is performed in this way, if either the interest level or similarity level is 0, the corrected value will also be 0. For example, the similarity between user A (post Q posted by user A) and user C is 0.9, but user C's interest level in user A is 0.0, so the result of multiplication is 0.0. As a result, user C will never be selected as a candidate no matter how high the similarity level is.

[0075] Therefore, when the similarity calculation unit 103 and the relationship calculation unit 105 respectively calculate the similarity and the relationship, a process of adding a surplus as shown in FIG. 19 is considered. FIG. 19 is a schematic diagram showing a process taking into consideration both the interest level and the similarity level. FIG. 19(a) is a diagram showing that the interest level of each user in user A is added with a surplus of 0.1. That is, 0.1 is added to each interest level. FIG. 19(b) is a diagram showing that the similarity level between user A and each user is added with a surplus of 0.1. In the diagram, 0.1 is added to each similarity level.

[0076] 19(c) is a diagram showing the similarity corrected based on the interest degree and the similarity degree. As shown in the figure, since none of the interest degrees and similarities are 0, the similarity degree after correction is also greater than 0. The selection unit 107 selects a user based on the similarity degree after correction.

[0077] Compared with the corrected similarity in Fig. 18(c), the corrected similarity between user A and user D in Fig. 19(c) is 0.1, and the recommendation ranking of user D is third. In Fig. 18(c), the corrected similarity for user C is 0, and so the recommendation ranking is fourth.

[0078] The selection unit 107 selects one or more users according to the recommendation order. For example, the selection unit 107 selects the users with the first and second recommendation orders. Alternatively, the selection unit 107 may select all users regardless of the recommendation order, and the notification unit 108 may notify according to the recommendation order. For example, a notification may be sent to the user with the first recommendation order, and if there is no reply within a predetermined time, the notification may be sent to the user with the next highest recommendation order.

[0079] Furthermore, the notification unit 108 may notify the user who posted the post Q of other users who are candidate users and their recommendation rankings.

[0080] Furthermore, the notification unit 108 may set the other users as targets for publishing the post Q by a user based on the recommendation order. The notification unit 108 may notify the QA posting service server 200 of the IDs or the like of users in a predetermined order, so that the QA posting service server 200 may make the post Q viewable only by users in the predetermined order, or may accept only answers from the users.

[0081] By thus weighting the similarity and interest levels, it is possible to carry out appropriate selection processing according to the value of the similarity or interest level.

[0082] In FIG. 19, the same value (0.1) is used for both interest and similarity as an increment, but the degree of correction can be adjusted by adjusting these values ​​as follows.

[0083] · Large interest bonus · Small similarity bonus: Candidates are selected with emphasis on similarity ·Small interest level · Large similarity level: Candidates are selected with an emphasis on interest level By giving a larger weight to the interest level than to the similarity level, it is possible to prevent only people with high relationships from being recommended as candidates.

[0084] Next, the effect of the user selection device 100 of the present disclosure will be described. In the user selection device 100 of the present disclosure, the similarity calculation unit 103 calculates the similarity between the post Q (posted content) posted by the user and the user information of the other user (the profile, posted content, etc. of the other user). Then, the relationship calculation unit 105 calculates the similarity and the relationship between the user and the other user. The selection unit 107 selects the other user as a candidate user based on the calculated relationship and similarity. With this configuration, it is possible to select a candidate user based on the relationship between the users in addition to the similarity of the posted content. The candidate user selected in consideration of the relationship is expected to reply to the post to one user.

[0085] The user selection device 100 includes a storage unit 104 that stores, for each user, basis information for calculating a relationship between the user and other users. The relationship calculation unit calculates the relationship between the user and other users based on the basis information. The basis information is information that indicates the directionality of the connection from the user to the other users.

[0086] For example, the storage unit 104 stores designated user information designated by the user as the basis information in advance. The designated user information is nomination information in this disclosure, and is information registered in advance by the user. The user may register other users who have similar skills to the user, other users who have similar values ​​or ways of thinking to the user, other users with whom the user frequently contacts for work, other users with whom the user frequently contacts outside of work, and the like.

[0087] The storage unit 104 also stores, as the basis information, a QA interaction history, which is answer history information indicating a posted Q (posted content) from the user and answers from other users. The storage unit 104 also stores, as the basis information, a profile viewing history indicating viewers of the user's profile.

[0088] With this configuration, the nomination information, Q&A exchange history, and profile viewing history indicate the relationships between users, and by using these, it is possible to determine those relationships.

[0089] In addition, when the relationship calculation unit 105 of the present disclosure calculates the relationship using at least two of the designated user information specified by the user, the QA interaction history showing the posted Q from the user and the answers from other users, and the user's profile browsing history, it may calculate the relationship by performing a weighting process on the connections from the user to the user in each of the nomination information, the QA interaction history, and the profile browsing history.

[0090] By carrying out weighting processing in this manner, it is possible to adjust the relationship between users as to which information should be emphasized.

[0091] In addition, in the present disclosure, in the user selection device 100, the similarity calculation unit 103 and the relationship calculation unit 105 may process the values ​​of the similarity and the relationship based on a predetermined adjustment value, and the correction unit 106 may correct the processed similarity using the relationship between users. These adjustment values ​​may be values ​​according to the similarity and the relationship, respectively. Although 0.1 is added uniformly in the above embodiment, the present invention is not limited to this.

[0092] This allows at least one of the similarity or relationship to be adjusted to a value other than 0 if it is 0, and even if the correction is made by multiplication, the correction of the similarity can be performed appropriately.

[0093] Furthermore, in the user selection device 100 of the present disclosure, when another user is selected as a candidate user, the selection unit 107 determines the recommendation ranking of the other user using the corrected similarity.

[0094] This allows the user to make selections based on appropriate similarities.

[0095] In the present disclosure, it is assumed that the one user and the other users belong to the same organization. For example, the user selection device 100 selects one or more colleagues from among several colleagues in the same company as the other users. When selecting a user from the same organization who is expected to reply to a post, the relationship between the users is particularly important. For this reason, the storage unit 104 stores, for example, the Q&A exchange history and profile viewing history between users in a user list in the company, as well as nomination information.

[0096] The user selection device 100 of the present disclosure further includes a notification unit 108 that notifies the user that a post has been made by the user. The selection unit 107 determines a recommendation order for each of the multiple other users, and the notification unit 108 performs notification processing for the multiple other users according to the recommendation order. For example, the notification unit 108 may perform notification processing in order of recommendation order at a predetermined time interval.

[0097] Furthermore, the notification unit 108 may notify the user who posted the post Q of other users who are candidate users and their recommendation rankings. This allows the user to know from which other users a reply will come.

[0098] Furthermore, based on the recommendation ranking of other users determined by the selection unit 107, the other users may be made publicly available to the one user's post Q.

[0099] That is, when a post Q is made by one user, only other users who are candidates may be allowed to view the posted content and respond to it.

[0100] The user selection device 100 notifies the QA posting service server 200 of the IDs of other users, so that the QA posting service server 200 can control whether or not to allow access based on the IDs.

[0101] The user selection device 100 of the present disclosure has the following configuration.

[0102] [1] a similarity calculation unit that calculates a similarity between content posted by a user and user information of another user; A relationship calculation unit that calculates a relationship between the user and another user; A selection unit that selects the other user as a candidate user based on the relationship and the similarity; A user selection device comprising:

[0103] [2] A storage unit is further provided for storing, for each of the users, basis information for calculating a relationship with another user, The relationship calculation unit calculates a relationship between the user and the other user based on the grounds information. The user selection device according to [1].

[0104] [3] The basis information is information having a direction of a connection from the user to other users. The user selection device according to [2].

[0105] [4] The storage unit stores, in advance, designated user information designated by the user as the basis information. The user selection device according to [3].

[0106] [5] The storage unit stores, as the basis information, answer history information indicating the content posted by the user and answers from the other users. The user selection device according to [3] or [4].

[0107] [6] The storage unit stores, as the basis information, browsing history information indicating a person who has viewed the profile of the user. A user selection device according to any one of [3] to [5].

[0108] [7] The relationship calculation unit, When calculating the relationship using at least two of designated user information designated by the user, answer history information indicating the content posted by the user and the answers from the other users, and browsing history information indicating the viewers of the user's profile, Calculating a relationship by performing a weighting process on the connection from the user to the previous user in each of the designated user information, the answer history information, and the browsing history information. The user selection device according to [2].

[0109] [8] a correction unit that corrects the similarity based on the relationship, The selection unit performs user selection using the corrected similarity. A user selection device according to any one of [1] to [7].

[0110] [9] The similarity calculation unit processes the similarity based on a predetermined adjustment value, The relationship calculation unit processes the relationship based on a predetermined adjustment value, The correction unit corrects the processed similarity using the processed relationship. [8] A user selection device according to the present invention.

[0111]

[10] The selection unit is determining a recommendation ranking of the other users based on the similarity and the relationship when the other users are selected as candidate users; A user selection device according to any one of [1] to [9].

[0112] The block diagrams used in the description of the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. The method of realizing each functional block is not particularly limited. That is, each functional block may be realized by using one device that is physically or logically combined, or may be realized by using two or more devices that are physically or logically separated and directly or indirectly connected (for example, by wire, wireless, etc.). The functional block may be realized by combining the one device or the multiple devices with software.

[0113] Functions include, but are not limited to, judgment, determination, judgment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocation, mapping, assignment, etc. For example, a functional block (component) that performs the function of transmission is called a transmitting unit or a transmitter. In either case, as described above, there is no particular limitation on the method of realization.

[0114] For example, the user selection device 100 according to an embodiment of the present disclosure may function as a computer that performs processing of the user selection method according to the present disclosure. Fig. 20 is a diagram showing an example of a hardware configuration of the user selection device 100 according to an embodiment of the present disclosure. The above-mentioned user selection device 100 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

[0115] In the following description, the term "apparatus" may be replaced with a circuit, a device, a unit, etc. The hardware configuration of the user selection device 100 may be configured to include one or more of the devices shown in the figure, or may be configured to exclude some of the devices.

[0116] Each function in the user selection device 100 is realized by loading a specific software (program) onto hardware such as a processor 1001 and a memory 1002, causing the processor 1001 to perform calculations, control communications via the communication device 1004, and control at least one of the reading and writing of data in the memory 1002 and the storage 1003.

[0117] The processor 1001, for example, operates an operating system to control the entire computer. The processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, the above-mentioned posted information characteristic vectorization unit 101, user information characteristic vectorization unit 102, similarity calculation unit 103, relationship calculation unit 105, correction unit 106, and selection unit 107 may be realized by the processor 1001.

[0118] Moreover, the processor 1001 reads out a program (program code), a software module, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002, and executes various processes according to the programs. As the programs, a program that causes a computer to execute at least a part of the operations described in the above-mentioned embodiment is used. For example, the posted information characteristic vectorization unit 101, the user information characteristic vectorization unit 102, the similarity calculation unit 103, the relationship calculation unit 105, the correction unit 106, and the selection unit 107 may be realized by a control program stored in the memory 1002 and operated in the processor 1001, and other functional blocks may be similarly realized. Although the above-mentioned various processes have been described as being executed by one processor 1001, they may be executed by two or more processors 1001 simultaneously or sequentially. The processor 1001 may be implemented by one or more chips. The programs may be transmitted from a network via an electric communication line.

[0119] The memory 1002 is a computer-readable recording medium, and may be configured by at least one of, for example, a Read Only Memory (ROM), an Erasable Programmable ROM (EPROM), an Electrically Erasable Programmable ROM (EEPROM), a Random Access Memory (RAM), and the like. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), and the like. The memory 1002 can store an executable program (program code), software module, and the like for implementing a user selection method according to an embodiment of the present disclosure.

[0120] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, and the like. Storage 1003 may be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other suitable medium including at least one of memory 1002 and storage 1003.

[0121] The communication device 1004 is hardware (transmission / reception device) for performing communication between computers via at least one of a wired network and a wireless network, and is also called, for example, a network device, a network controller, a network card, a communication module, etc. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc., in order to realize at least one of, for example, Frequency Division Duplex (FDD) and Time Division Duplex (TDD). For example, the above-mentioned notification unit 108 may be realized by the communication device 1004. The communication device 1004 may be implemented with a transmission unit and a reception unit that are physically or logically separated.

[0122] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that performs output to the outside. The input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0123] In addition, each device such as the processor 1001 and the memory 1002 is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0124] Furthermore, the user selection device 100 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), etc., and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0125] The notification of information is not limited to the aspects / embodiments described in the present disclosure, and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI), Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB), System Information Block (SIB))), other signals, or a combination thereof. In addition, the RRC signaling may be called an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.

[0126] The order of the steps, sequences, flow charts, etc. of each aspect / embodiment described in this disclosure may be changed unless inconsistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0127] The input and output information may be stored in a specific location (e.g., memory) or may be managed using a management table. The input and output information may be overwritten, updated, or added to. The output information may be deleted. The input information may be transmitted to another device.

[0128] The determination may be based on a value represented by a single bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0129] Each aspect / embodiment described in this disclosure may be used alone, in combination, or switched according to execution. In addition, notification of specific information (e.g., notification that "X is the case") is not limited to being done explicitly, but may be done implicitly (e.g., not notifying the specific information).

[0130] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described in the present disclosure. The present disclosure can be implemented as modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0131] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0132] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if the software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave, etc.), then these wired and / or wireless technologies are included within the definition of transmission media.

[0133] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, the data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0134] In addition, the terms described in this disclosure and the terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings. For example, at least one of the channel and the symbol may be a signal (signaling). Also, the signal may be a message. Also, the component carrier (CC) may be called a carrier frequency, a cell, a frequency carrier, etc.

[0135] In addition, the information, parameters, etc. described in the present disclosure may be represented using absolute values, may be represented using relative values ​​from a predetermined value, or may be represented using other corresponding information. For example, a radio resource may be indicated by an index.

[0136] The names used for the above-mentioned parameters are not limiting in any way. Moreover, the formulas etc. using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (e.g., PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not limiting in any way.

[0137] In this disclosure, the terms "Mobile Station (MS)", "user terminal", "User Equipment (UE)", "terminal", etc. may be used interchangeably.

[0138] A mobile station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0139] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, search, inquiry (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in a memory), and the like. In addition, "judgment" and "decision" can include considering resolving, selecting, choosing, establishing, comparing, etc., to be a "judgment" or "decision." In other words, "judgment" and "decision" can include considering some action to be a "judgment" or "decision." In addition, "judgment" can be interpreted as "assuming," "expecting," "considering," etc.

[0140] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using at least one of one or more wires, cables, and printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0141] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0142] When "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Further, when used in this disclosure, the term "or" is not intended to be an exclusive or.

[0143] In this disclosure, where articles have been added due to translation, such as, for example, a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0144] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different." [Explanation of symbols]

[0145] 100...user selection device, 200...QA posting service server, 300...user terminal, 101...posted information feature vectorization unit, 102...user information feature vectorization unit, 103...similarity calculation unit, 104...memory unit, 105...relationship calculation unit, 106...correction unit, 107...selection unit, 108...notification unit.

Claims

1. a similarity calculation unit that calculates a similarity between content posted by a user and user information of another user; A relationship calculation unit that calculates a relationship between the user and another user; A selection unit that selects the other user as a candidate user based on the relationship and the similarity; A user selection device comprising:

2. A storage unit is further provided for storing, for each of the users, basis information for calculating a relationship with another user, The relationship calculation unit calculates a relationship between the user and the other user based on the grounds information. The user selection device of claim 1 .

3. The basis information is information having a direction of a connection from the user to other users. The user selection device of claim 2.

4. The storage unit stores, in advance, designated user information designated by the user as the basis information.

4. The user selection device of claim 3.

5. The storage unit stores, as the basis information, answer history information indicating the content posted by the user and answers from the other users.

4. The user selection device of claim 3.

6. The storage unit stores, as the basis information, browsing history information indicating a person who has viewed the profile of the user.

4. The user selection device of claim 3.

7. The relationship calculation unit, When calculating the relationship using at least two of designated user information designated by the user, answer history information indicating the content posted by the user and answers from the other users, and browsing history information indicating the viewers of the user's profile, Calculating a relationship by performing a weighting process on the connection from the user to the previous user in each of the designated user information, the answer history information, and the browsing history information. The user selection device of claim 2.

8. a correction unit that corrects the similarity based on the relationship, The selection unit performs user selection using the corrected similarity. The user selection device of claim 1 .

9. The similarity calculation unit processes the similarity based on a predetermined adjustment value, The relationship calculation unit processes the relationship based on a predetermined adjustment value, The correction unit corrects the processed similarity using the processed relationship.

9. A user selection device according to claim 8.

10. The selection unit is determining a recommendation ranking of the other users based on the similarity and the relationship when the other users are selected as candidate users; The user selection device of claim 1 .

Images