Evaluation equipment

Abstract

To provide an evaluation device that performs evaluation without relying on a user's active input operation using an evaluation object.SOLUTION: In a server device 10 serving as an evaluation device, an acquisition unit 11 acquires a user's conversation after using a facility to be evaluated, and a keyword extraction unit 12 extracts keywords from the acquired conversation according to predetermined rules. Next, a comment extraction unit 13 extracts a comment whose degree of matching with the keywords satisfies a predetermined standard from the server device in which public comments about the facility to be evaluated and evaluation values associated with the comments are stored. Then, a determination unit 14 determines an evaluation value for the facility to be evaluated by the user based on the evaluation value associated with the extracted comment.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an evaluation device that evaluates an evaluation target such as a visited facility. [Background technology]

[0002] For example, when a user visits a facility or the like, it is common to obtain information regarding the evaluation of the facility and use this evaluation information to provide various services.

[0003] Such evaluation information is typically provided actively by users via evaluation sites, etc., after they return home from the facility, but many users find it cumbersome to have to go to the trouble of filling out evaluations, etc., and proposals have been made to make it easier to obtain user evaluations (see, for example, Patent Document 1).

[0004] Patent document 1 describes that when a user visits a facility and displays an image through a viewing UI, if a predetermined condition is met that there is no evaluation information for the facility corresponding to the location information associated with the image, an evaluation UI is automatically provided, thereby easily providing facility users with an evaluation user interface without relying on active operations such as the facility users having to take the time to access a website to input evaluation information. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2020-119274 Summary of the Invention [Problem to be solved by the invention]

[0006] Although the method described in Patent Document 1 does eliminate the need for facility users to take the time to actively access a website to input evaluation information, users still need to input their evaluations, which can be cumbersome for users.

[0007] One example of the problem that the present invention aims to solve is to enable evaluation without relying on active input operations by the user using the evaluation object, or to reduce the burden on the user for input operations. [Means for solving the problem]

[0008] In order to solve the above problem, the invention described in claim 1 is characterized by comprising an acquisition unit that acquires comments made by users after using an evaluation object; a keyword extraction unit that extracts keywords from the acquired comments in accordance with predetermined rules; a comment extraction unit that extracts comments whose degree of match with the keywords meets a predetermined standard from a storage unit that accumulates public comments about the evaluation object and evaluation values ​​associated with the comments; and a determination unit that determines the user's evaluation value for the evaluation object based on the evaluation value associated with the extracted comment.

[0009] The invention described in claim 7 is an evaluation method executed by an evaluation device that evaluates a user's evaluation object, and is characterized by including: an identification step of identifying the evaluation object; an acquisition step of acquiring comments made by the user after using the identified evaluation object; a keyword extraction step of extracting keywords from the acquired comments in accordance with predetermined rules; a comment extraction step of extracting comments whose degree of coincidence with the keywords meets a predetermined standard from a storage unit that accumulates public comments about the evaluation object and evaluation values ​​associated with the comments; and a determination step of determining the user's evaluation value for the evaluation object based on the evaluation value associated with the extracted comment.

[0010] The invention as set forth in claim 8 is characterized in that the evaluation method as set forth in claim 7 is executed by a computer as an evaluation program.

[0011] The invention as set forth in claim 9 is characterized in that the evaluation program as set forth in claim 8 is stored in a computer-readable storage medium. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a schematic configuration diagram of a system including an evaluation device according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a functional configuration diagram of the server device shown in FIG. [Figure 3] FIG. 2 is a functional configuration diagram of the vehicle-mounted device shown in FIG. [Figure 4] 2 is a flowchart of the operation of the server device shown in FIG. [Figure 5] FIG. 10 is a schematic configuration diagram of a system including an evaluation device according to a second embodiment of the present invention. [Figure 6] 6 is a flowchart of the operation of the server device shown in FIG. 5. DETAILED DESCRIPTION OF THE INVENTION

[0013] An evaluation device according to one embodiment of the present invention will now be described. In the evaluation device according to one embodiment of the present invention, an acquisition unit acquires comments made by users after using an evaluation object, and a keyword extraction unit extracts keywords from the acquired comments according to a predetermined rule. Next, a comment extraction unit extracts comments whose degree of match with the keywords meets a predetermined standard from a storage unit that accumulates public comments about the evaluation object and the evaluation values ​​associated with the comments. A determination unit then determines the user's evaluation value for the evaluation object based on the evaluation values ​​associated with the extracted comments. This allows the user to determine their own evaluation based on the evaluation values ​​of other users whose degree of match with the keywords extracted from the user's comments meets a predetermined standard. This saves the user the trouble of inputting an evaluation. Therefore, evaluations can be made without relying on active input operations by the user who has used the evaluation object, thereby reducing the burden on the user's input operations.

[0014] The determination unit may also determine the evaluation value associated with the comment extracted by the comment extraction unit as the user's evaluation value for the evaluation object. In this way, the evaluation values ​​of other users can be used as the user's own evaluation value, eliminating the need to calculate evaluation values ​​and simplifying the evaluation process.

[0015] Furthermore, the determination unit may determine, as the user's evaluation value for the evaluation object, the evaluation value associated with a comment that has a high degree of coincidence with the keyword extracted by the keyword extraction unit, among the comments extracted by the comment extraction unit. By doing so, the evaluation value of a comment whose opinion is closer to the user's own opinion can be set as the user's own evaluation value. Therefore, the accuracy of the evaluation value can be improved.

[0016] The acquisition unit may also acquire conversations between users after using the evaluation object. In this way, the evaluation of the evaluation object is automatically performed simply by verbally conversing with the user without the user being aware of inputting an evaluation.

[0017] The acquisition unit may also acquire conversations between users while they are in the vehicle. In this way, the evaluation of the evaluation object can be automatically performed based on conversations between users while they are traveling in the vehicle.

[0018] The evaluation object may also be a facility that the user has used, allowing the user to evaluate a facility such as a restaurant that the user has used.

[0019] In addition, an evaluation method according to one embodiment of the present invention includes an acquisition step of acquiring user comments after using an evaluation object, and a keyword extraction step of extracting keywords from the acquired comments according to a predetermined rule. Next, a comment extraction step of extracting comments whose degree of match with the keywords meets a predetermined standard from a storage unit that stores public comments about the evaluation object and the evaluation values ​​associated with the comments. Then, a determination step of determining the user's evaluation value for the evaluation object based on the evaluation values ​​associated with the extracted comments. This allows the user to determine their own evaluation based on the evaluation values ​​of other users whose degree of match with the keywords extracted from the user's comments meets a predetermined standard. This saves the user the trouble of inputting an evaluation. Therefore, evaluations can be made without relying on active input operations by the user who has used the evaluation object, thereby reducing the burden on the user's input operations.

[0020] The above-described evaluation method may also be executed by a computer as an evaluation program. This allows the computer to determine the user's own evaluation based on the evaluation values ​​of other users whose degree of similarity with keywords extracted from the user's comments meets a predetermined standard. This saves the user the trouble of inputting an evaluation. Therefore, evaluations can be made without relying on the user's active input operation using the evaluation target.

[0021] The evaluation program may be stored in a computer-readable storage medium, which allows the program to be distributed as a standalone program rather than being incorporated into a device, and allows for easy version upgrades. [Example]

[0022] An evaluation device according to a first embodiment of the present invention will be described with reference to Figures 1 to 4. Figure 1 is a schematic diagram of a system including an evaluation device according to this embodiment.

[0023] The system 100 includes a server device 10, an in-vehicle device 20, and a server device 30. The server device 10 constitutes an evaluation device according to this embodiment. The server device 10 is capable of communicating with the in-vehicle device 20 and the server device 30 (described later) via a network N such as the Internet.

[0024] The server device 10 is a well-known server computer, and includes a microprocessor including a CPU (Central Processing Unit), a memory, etc., a storage device such as a hard disk, a network interface, and the like.

[0025] 2 shows the functional configuration of the server device 10. The server device 10 includes an acquisition unit 11, a keyword extraction unit 12, a comment extraction unit 13, and a determination unit 14.

[0026] The acquisition unit 11 acquires user comments after using the evaluation object. The acquisition unit 11 functions as, for example, a network interface. In this embodiment, the evaluation object is a facility that the user visited as a destination or a stopover, as will be described later. The acquisition unit 11 acquires the user comments acquired by the in-vehicle device 20 via the network N.

[0027] The keyword extraction unit 12 extracts keywords from the utterances acquired by the acquisition unit 11 in accordance with predetermined rules. The utterances may be extracted using well-known natural language analysis of speech recognition or the like. In this embodiment, keywords to be extracted are set in advance, and the set keywords are extracted from the utterances after speech recognition. These set keywords may be set for each facility. For example, for a restaurant, keywords such as "XX is delicious," "The staff are kind," and "The interior of the restaurant is stylish" may be set (not limited to words, but may also be phrases). Keywords for each facility may also be set by extracting them from evaluation comments accumulated in the server device 30 described below.

[0028] The comment extraction unit 13 extracts comments whose degree of match with the keywords extracted by the keyword extraction unit 12 meets a predetermined criterion from the comments stored in the server device 30. The server device 30 stores ratings given by multiple other users for facilities and comments corresponding to each rating, as will be described later, and extracts comments whose degree of match with the keywords meets a predetermined criterion. For example, if multiple keywords are extracted, the predetermined criterion can be a criterion that includes all of the keywords or more than half of the keywords.

[0029] The determination unit 14 determines the user's evaluation value for the evaluation object based on the evaluation value associated with the extracted comment. In this embodiment, the determination unit 14 determines the evaluation value associated with the extracted comment as the user's evaluation value for the evaluation object. For example, if the evaluation value associated with the extracted comment is 3 out of 5, the user's evaluation value is 3. In other words, the evaluation value associated with opinions similar to the user's is used as the user's evaluation value. Note that if multiple comments are extracted, the average of those evaluation values ​​may be used as the user's evaluation value, or the mode or median may be used as the user's evaluation value.

[0030] Furthermore, the determination unit 14 may determine, as the evaluation value for the evaluation target of the user, the evaluation value associated with a comment that has a high degree of match with the keyword among the extracted comments. For example, if the criterion extracted by the comment extraction unit 13 is that a comment containing more than half of the keyword is included, and if a comment containing half of the keyword and a comment containing all of the keyword are extracted, the determination unit 14 adopts the evaluation value associated with the comment containing all of the keyword.

[0031] The keyword extraction unit 12, the comment extraction unit 13, and the determination unit 14 function as computer programs that run on the microprocessor of the server device 10.

[0032] The in-vehicle device 20 is a device installed in the vehicle V or a device used in the vehicle V. Specific examples of the in-vehicle device 20 include a car navigation device and a drive recorder. As shown in FIG. 3 , the in-vehicle device 20 includes a microphone 21, a camera 22, a control unit 23, and a communication unit 24. Note that the in-vehicle device 20 is not limited to a device fixed inside the vehicle, and may be a portable device carried by a passenger.

[0033] The microphone 21 picks up (acquires) the conversation of users and the like aboard the vehicle V. The microphone 21 converts the acquired conversation into an electrical signal and outputs it to the control unit 23.

[0034] The camera 22 captures images of the interior and exterior of the vehicle V. The camera 22 can be used for user authentication, for example, by capturing an image of a user inside the vehicle.

[0035] The control unit 23 is configured with a microprocessor having a CPU, memory, etc. The control unit 23 is responsible for overall control of the in-vehicle device 20. In this embodiment, the control unit 23 transmits an audio signal acquired by the microphone 21 to the server device 10 via the communication unit 24. Furthermore, if the control unit 23 is a navigation device, it searches for a route from the current location to a set destination and provides guidance along the searched route. If the control unit 23 is a drive recorder, it stores images captured by the camera 22 in a storage device (not shown) and transmits them to the server device 10 via the communication unit 24.

[0036] The communication unit 24 communicates wirelessly with the network N. The communication unit 24 transmits the above-mentioned voice information of the conversation and the like to the server device 10.

[0037] The server device 30 is a well-known server computer similar to the server device 10, and includes a microprocessor including a CPU (Central Processing Unit), a memory, etc., a storage device such as a hard disk, a network interface, etc.

[0038] The server device 30 stores public comments about evaluation objects and evaluation values ​​associated with the comments. For example, it stores user comments about a certain facility and evaluation values ​​associated with the user comments. That is, the server device 30 functions as a storage unit. Furthermore, the comments and evaluation values ​​stored in the server device 30 do not have to be collected solely for use by the server device 10. The server device 30 allows an unspecified number of people to save new comments and evaluation values ​​via the Internet N or the like (this may be a member-only service, allowing a specified number of people to save new comments and evaluation values). Furthermore, the comments and the like stored in the server device 30 may be viewable by third parties via the Internet N or the like.

[0039] Next, the operation (evaluation method) of the server device 10 (evaluation device) configured as described above will be described with reference to the flowchart in Fig. 4. The flowchart in Fig. 4 can be configured as a program executed by a computer equipped with a CPU or the like, thereby forming an evaluation program.

[0040] First, the comment extraction unit 13 identifies a facility to be evaluated (step S11). For example, if the in-vehicle device 20 is a navigation device, the facility may be identified by acquiring information on the set destination or route points. Alternatively, the server device 10 or another server device may have POI (Point of Interest) information, and the facility may be identified by extracting the facility name contained in the POI information around the current location from the acquired conversation information. Note that when the facility is identified from the acquired conversation information, the order of the facility identification is reversed with the next step S2.

[0041] Next, the acquisition unit 11 acquires conversation information (step S12). That is, the conversation information is acquired from the vehicle-mounted device 20 via the Internet N. Then, the keyword extraction unit 12 extracts keywords (step S13). That is, keywords are extracted from the utterances acquired by the acquisition unit 11 according to predetermined rules.

[0042] Next, the comment extraction unit 13 acquires comments and the like about the facility identified in step S11 from the server device 30, and extracts comments that satisfy a predetermined criterion from the acquired comments and the like (step S14). That is, from the comments accumulated in the server device 30, comments whose degree of coincidence with the keywords extracted by the keyword extraction unit 12 satisfies a predetermined criterion are extracted. Then, the determination unit 14 determines an evaluation value for the facility to be evaluated (step S15). That is, the determination unit 14 determines an evaluation value for the user's evaluation target based on the evaluation value associated with the extracted comment.

[0043] As is clear from the above description, step S12 functions as an acquisition step, step S13 functions as a keyword extraction step, step S14 functions as a comment extraction step, and step S15 functions as a determination step.

[0044] Here, a specific example of the operation of the server device 10, which is the evaluation device according to this embodiment, will be described. Suppose that on the way back from Restaurant X, the destination of a vehicle V equipped with an on-board device 20, a user A, a passenger of the vehicle V, utters, "The hamburger was delicious, oozing with juice." This utterance is picked up by the microphone 21 of the on-board device 20 and transmitted from the on-board device 20 to the server device 10. Furthermore, since the destination was Restaurant X, Restaurant X is identified as the facility to be evaluated.

[0045] Then, from the comment "The hamburger steak was juicy and delicious," keywords such as "hamburger steak," "juicy," and "delicious" are extracted, and comments containing many of these keywords are extracted from the server device 30. As a result of the extraction, let us assume that the comment "The juicy hamburger steak was delicious" is extracted, and the evaluation value corresponding to this comment is 4 points out of 5. In this case, the evaluation value of Restaurant X for User A is 4 points.

[0046] This evaluation value of restaurant X by user A may be transmitted, for example, from server device 10 to server device 30, and reflected in the evaluation of restaurant X. Alternatively, it may be transmitted from server device 10 to vehicle-mounted device 20, and the vehicle-mounted device 20 may notify user A of the evaluation value. At this time, furthermore, a correction operation from user A to the evaluation value may be accepted.

[0047] According to this embodiment, the server device 10 has an acquisition unit 11 that acquires conversations between users after using a facility to be rated, and a keyword extraction unit 12 that extracts keywords from the acquired conversations according to a predetermined rule. Next, the comment extraction unit 13 extracts comments whose degree of match with keywords meets a predetermined standard from the server device 30, which stores public comments about the facility to be rated and the evaluation values ​​associated with the comments. The determination unit 14 then determines the user's evaluation value for the facility to be rated based on the evaluation values ​​associated with the extracted comments. This allows the user to determine their own evaluation based on the evaluation values ​​of other users whose degree of match with keywords extracted from the user's comments meets a predetermined standard. This eliminates the need for the user to input an evaluation. Therefore, evaluations can be made without relying on the user's active input operation after using the facility to be rated. Furthermore, the burden on the user's input operation can be reduced.

[0048] The determination unit also determines the evaluation value associated with the comment extracted by the comment extraction unit 13 as the user's evaluation value for the facility being rated. In this way, the evaluation values ​​of other users can be used as the user's own evaluation value, eliminating the need to calculate evaluation values ​​and simplifying the evaluation process.

[0049] Furthermore, the determination unit determines, as the user's evaluation value for the facility being rated, the evaluation value associated with a comment that has a high degree of coincidence with the keywords extracted by the keyword extraction unit 12, among the comments extracted by the comment extraction unit 13. By doing so, the evaluation value of a comment whose opinion is closer to the user's own opinion can be set as the user's own evaluation value. This makes it possible to increase the accuracy of the evaluation value.

[0050] The acquisition unit 11 also acquires conversations between users after using the facility to be evaluated. In this way, the user can automatically evaluate the facility to be evaluated by simply speaking orally, without being aware of inputting an evaluation.

[0051] Furthermore, the acquisition unit 11 acquires conversation information of the user inside the vehicle V. In this way, the evaluation of the facility to be evaluated can be automatically performed from the conversation while traveling in the vehicle V.

[0052] In the above-described embodiment, the server device 10 includes the acquisition unit 11, the keyword extraction unit 12, the comment extraction unit 13, and the determination unit 14, but the present invention is not limited to this configuration. For example, the acquisition unit 11 may be implemented by the vehicle-mounted device 20. Furthermore, the vehicle-mounted device 20 may have the functions of the acquisition unit 11 and the keyword extraction unit 12. Furthermore, the vehicle-mounted device 20 may function as an evaluation device.

[0053] Furthermore, if conversations within the vehicle cannot be confirmed or if there is only one occupant, the vehicle-mounted device 20 may make utterances to start a conversation. For example, on the way home from Restaurant X, the vehicle-mounted device 20 may make utterances such as "What was Restaurant X like?" and "Was the food delicious?" The vehicle-mounted device 20 acquires responses to these utterances and transmits them to the server device 10, which then associates the responses with comments from the public and determines an evaluation as described above. [Example]

[0054] Next, an evaluation device according to a second embodiment of the present invention will be described with reference to Figures 5 and 6. Note that the same parts as those in the first embodiment described above are given the same reference numerals and their description will be omitted.

[0055] 5 is a schematic diagram of a system including an evaluation device according to this embodiment. The system 100A includes a server device 10A, a server device 30, a smartphone 40, and an SNS server 50. The basic configuration of the server device 10A is the same as that of the first embodiment.

[0056] In this embodiment, the system 100A includes a smartphone 40 instead of the in-vehicle device 20. As is well known, the smartphone 40 is a multi-function mobile phone that can be customized to suit the user's preferences by later adding software (applications) or the like. Note that, in this embodiment, a smartphone is described as one form of terminal, but other mobile terminals such as a tablet PC may also be used. Alternatively, the smartphone 40 may be a PC (personal computer) installed at home or the like.

[0057] In this embodiment, evaluations can be made not only for facilities visited by vehicle V, but also for facilities visited by other means such as train or walking, and for objects such as home appliances and automobiles. Furthermore, the comments acquired by server device 10A are not conversations inside the vehicle but are text information, and specifically, may be posts on a social networking service (SNS) or a messenger app other than the evaluation site (server device 30).

[0058] 5, an SNS server 50 is added. The SNS server 50 is a server device that provides an SNS service, and posts and views can be made to the SNS server 50 from the smartphone 40. The server device 10A acquires posts from users from the SNS server 50 and evaluates the evaluation object in accordance with the posts, similar to the first embodiment.

[0059] Next, the operation (evaluation method) of the server device 10A (evaluation device) according to this embodiment will be described with reference to the flowchart of Fig. 6. The basic operation of the flowchart of Fig. 6 is similar to that of the flowchart of Fig. 4. In the flowchart of Fig. 6, the step of identifying a facility is included in the step of acquiring a conversation (step S12A).

[0060] In step S12A, the acquisition unit 11 acquires comment information. As described above, this comment information is acquired from the SNS server 50 in the form of user posts. Then, the evaluation target is identified from the posts. The evaluation target can be identified by determining whether a proper noun (not limited to names, but also including company names, model names, and model numbers) in the post is included in the evaluation target of the server device 30.

[0061] For example, if a post on an SNS describes the usability of a device with model number A001 from company A, "company A" and "A001" are extracted, and a search is performed to see if any comments or the like have been stored on the server device 30. If a comment is found as a result of the search, it is deemed to be the subject of evaluation, and the same processing as in the first embodiment is performed on the found comment.

[0062] According to this embodiment, the evaluation target can include things other than facilities, thereby broadening the scope of application. Furthermore, since the utterances can be text, a microphone is not required, and there is no need to capture conversations in real time. Furthermore, when inputting text, the user can automatically obtain an evaluation value by simply writing it on a site or the like that they normally use, without being aware of the evaluation. This eliminates the need for the user to take the time to actively access a website to input evaluation information. This reduces the burden on the user's input operations.

[0063] The present invention is not limited to the above-described embodiments. In other words, a person skilled in the art can implement various modifications in accordance with conventionally known knowledge without departing from the gist of the present invention. As long as such modifications still include the evaluation device of the present invention, they are of course included in the scope of the present invention. [Explanation of symbols]

[0064] 10 Server device (evaluation device) 11 Acquisition Department 12 Keyword extraction section 13 Comment Extraction 14 Decision Section 20 Onboard equipment 30 Server device (storage unit) 40 Smartphone 50 SNS servers

Claims

1. an acquisition unit that acquires comments from users after using the evaluation object; a keyword extraction unit that extracts keywords from the acquired comments in accordance with a predetermined rule; a comment extraction unit that extracts comments whose degree of coincidence with the keyword satisfies a predetermined standard from a storage unit that stores public comments about the evaluation object and evaluation values ​​associated with the comments; a determination unit that determines an evaluation value of the user for the evaluation object based on the evaluation value associated with the extracted comment; An evaluation device comprising:

2. The evaluation device according to claim 1 , wherein the determination unit determines an evaluation value associated with the extracted comment as the user's evaluation value for the evaluation target.

3. the determination unit determines, from among the extracted comments, an evaluation value associated with a comment that has a high degree of match with the keyword as the user's evaluation value for the evaluation target; 3. The evaluation device according to claim 2.

4. The evaluation device according to claim 1 , wherein the acquisition unit acquires a conversation of the user after using the evaluation object.

5. The evaluation device according to claim 4 , wherein the acquisition unit acquires conversations of the user in a vehicle.

6. 6. The evaluation device according to claim 1, wherein the evaluation object is a facility used by the user.

7. An evaluation method executed by an evaluation device that evaluates an evaluation target of a user, comprising: an acquisition step of acquiring comments from the user after using the evaluation object; a keyword extraction step of extracting keywords from the acquired comments in accordance with a predetermined rule; a comment extraction step of extracting comments whose degree of coincidence with the keyword satisfies a predetermined standard from a storage unit that stores public comments about the evaluation object and evaluation values ​​associated with the comments; a determination step of determining an evaluation value of the user for the evaluation object based on the evaluation value associated with the extracted comment; An evaluation method comprising:

8. An evaluation program for causing a computer to execute the evaluation method according to claim 7.

9. 9. A computer-readable recording medium storing the evaluation program according to claim 8.

Images