Recommended method

Abstract

To improve a rule related to recommendation.SOLUTION: A method for recommendation is a method of recommending a stop-off point to a user based on a predetermined rule. The method for recommendation includes: a reception step of receiving input of feedback information from a user, showing whether the recommendation of a stop-off point is appropriate; a determination step of determining whether the feedback information satisfies an improvement condition related to a predetermined rule; and an update step of updating the predetermined rule on the basis of the feedback information in a case where the feedback information has been determined to satisfy the improvement condition.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to the technical field of a recommendation method for recommending stopping places.

Background Art

[0002] As this type of method, for example, a method has been proposed in which the mental load of a user based on the user's movement history is used as a search condition to determine a proposed destination (see Patent Document 1). In addition, Patent Documents 2 to 5 are cited as prior art documents related to the present invention.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Patent Document 5

Summary of the Invention

Problems to be Solved by the Invention

[0004] For example, in the technique described in Patent Document 1, there is no configuration for using feedback from a user regarding a proposed destination. Therefore, the technique described in Patent Document 1 has a technical problem that even when a user determines that the proposed destination as a proposal result is inappropriate, the proposal result cannot be improved.

[0005] The present invention has been made in view of the above problems, and an object thereof is to provide a recommendation method capable of improving rules related to recommendations.

Means for Solving the Problems

[0006] A recommendation method according to one aspect of the present invention is: The navigation system installed in the vehicle, Based on the prescribed rules , boarding the aforementioned vehicle A recommendation method that suggests places to visit to the user, After a second image, different from the first image, is displayed on the navigation device in place of the first image which includes one or more recommended stops, The aforementioned user for the navigation device Please provide feedback information indicating whether the recommendations for stopover locations were appropriate. The aforementioned navigation device The acceptance process, The aforementioned navigation device A determination step of determining whether the feedback information satisfies the improvement conditions related to the predetermined rule, and if it is determined that the feedback information satisfies the improvement conditions, The aforementioned navigation device This includes an update step of updating the predetermined rule based on the feedback information. [Brief explanation of the drawing]

[0007] [Figure 1] Block diagram showing the configuration of a vehicle according to the embodiment. [Figure 2] This is a flowchart showing the operation of the navigation device according to the embodiment. [Figure 3] This figure shows an example of a displayed image. [Figure 4] This figure shows other examples of the displayed image. [Figure 5] This figure shows other examples of the displayed image. [Figure 6] This is a flowchart showing the feedback operation according to the embodiment. [Modes for carrying out the invention]

[0008] Embodiments of the recommendation method will be described with reference to Figures 1 to 6.

[0009] (Vehicle configuration) The configuration of the vehicle according to this embodiment will be described with reference to Figure 1. In Figure 1, the vehicle 1 includes an external communication device 11, a position detection device 12, an on-board camera 13, an on-board sensor 14, and a navigation device 20. The vehicle 1 may be a vehicle with an autonomous driving function.

[0010] The external communication device 11 is a device having wireless communication capabilities. The external communication device 11 is connected to a network via a wireless base station. The position detection device 12 is a device that detects the position of vehicle 1. The position detection device 12 may be, for example, a GPS (Global Positioning System) receiver.

[0011] The on-board camera 13 is not limited to a camera that images the area around the vehicle 1 (in other words, the area outside the vehicle 1), but may also be a camera that images the interior of the vehicle 1 (in other words, the interior of the vehicle 1). The on-board camera 13 may include two or more cameras. In this case, the on-board camera 13 may include a camera that images the area around the vehicle 1 and a camera that images the interior of the vehicle 1. The on-board sensor 14 may include, for example, at least one of a speed sensor, an acceleration sensor, a yaw rate sensor, a steering angle sensor, a temperature sensor, a humidity sensor, and a rainfall sensor.

[0012] The navigation device 20 comprises a control unit 21, a storage unit 22, an in-vehicle communication interface (I / F) 23, a display unit 24, and an operation unit 25. The control unit 21 may include one or more processors that execute a predetermined computer program. The storage unit 22 may be a storage medium that stores a predetermined computer program executed in the control unit 21. The in-vehicle communication I / F 23 may be a communication interface circuit for the navigation device 20 to communicate with other in-vehicle equipment in the vehicle 1 via an in-vehicle network.

[0013] The display unit 24 may be, for example, a liquid crystal display or an organic EL (Electro-Luminesence) display. The operation unit 25 may be at least one of, for example, a touch panel, a mouse, a keyboard, and operation buttons. Incidentally, the display unit 24 and the operation unit 25 may be integrated as a touch display.

[0014] (Operation of the Navigation Device) The operation of the navigation device 20 will be described with reference to the flowchart of FIG. 2. First, an overview of the operation of the navigation device 20 will be described. Then, the operation of the navigation device 20 will be described with specific examples.

[0015] In FIG. 2, the control unit 21 of the navigation device 20 selects feature amounts for recommending stopping places and calculates weighting parameters for the selected feature amounts (step S101). The feature amounts may include, for example, at least one of the current time, season, weather, vehicle data, road information, business hours, floor area, parking lot information, type of stopping place, and popularity. Incidentally, the popularity may be determined based on at least one of the number of visitors and the number of visits of each person. The control unit 21 may calculate the weighting parameters by performing a predetermined filtering process on the categorical feature amounts. The control unit 21 may calculate the weighting parameters by performing, for example, Laplace smoothing on the numerical feature amounts.

[0016] Next, the control unit 21 acquires selection criteria (step S102). For example, the control unit 21 may change the selection criteria based on at least one of the type of the recommended stopover location and the feature amount selected in the process of step S101. The selection criteria may include a calculation formula for merging the feature amounts selected in the process of step S101. The selection criteria may include the number of stopover locations recommended to the user of the vehicle 1. The selection criteria may include an exclusion filter for determining stopover locations to be excluded. Note that the selection criteria may be a learning model that outputs a stopover location recommended to the user when a feature amount is input. Since this learning model is used for recommending a stopover location, it may also be referred to as a recommendation model.

[0017] Next, the control unit 21 generates a recommendation result and a reason for recommendation (step S103). For example, the control unit 21 may calculate the score of each of a plurality of stopover location candidates using the calculation formula included in the selection criteria. The control unit 21 may select one or more stopover locations to be recommended to the user from the plurality of stopover location candidates based on the calculated scores. The control unit 21 may generate information indicating the selected one or more stopover locations as the above-mentioned recommendation result. For example, the control unit 21 may extract one or more feature amounts that have a large influence on the score among the plurality of feature amounts related to the selected one or more stopover locations. The control unit 21 may generate the above-mentioned reason for recommendation based on the extracted one or more feature amounts.

[0018] The control unit 21 determines the display timing of the recommendation result and the reason for recommendation (step S104). Note that the process of step S104 may be performed in parallel with the processes of steps S101 to S103. For example, the control unit 21 may determine a timing that does not affect the running of the vehicle 1 as the display timing. The timing that does not affect the running of the vehicle 1 may include at least one of when the position of the shift lever is "P" and when the vehicle 1 is stopped.

[0019] The control unit 21 controls the display unit 24 so that the recommendation result and recommendation reason are displayed at the display timing determined in step S104 (step S105). The control unit 21 also controls the display unit 24 so that the evaluation content is displayed. "Evaluation content" may refer to the user's evaluation items for the recommendation result. Specific examples of "recommendation result and recommendation reason" and "evaluation content" will be described later. The display timing of "recommendation result and recommendation reason" and the display timing of "evaluation content" may be different or the same.

[0020] The control unit 21 determines whether or not there is user feedback (step S106). If it is determined in step S106 that there is no user feedback (step S106: Yes), the operation shown in Figure 2 is terminated. After a first predetermined time has elapsed since the operation shown in Figure 2 was terminated, the process in step S101 may be performed. In other words, the operation shown in Figure 2 may be repeated at a cycle corresponding to the first predetermined time.

[0021] If it is determined that user feedback is available during the processing in step S106 (step S106: Yes), the control unit 21 improves the recommendation logic for stopover locations based on the feedback (step S107). Specific examples of how the recommendation logic can be improved will be described later. Improving the recommendation logic may mean improving at least some of the selection criteria mentioned above.

[0022] (First example) As a first specific example of the operation of the navigation device 20, the operation of recommending places to stop for leisure activities will be explained with reference to Figures 3 to 6 in addition to Figure 2. Places to stop for leisure activities may include at least one of restaurants and tourist attractions.

[0023] For example, the control unit 21 may acquire visit history data of at least one of restaurants and tourist attractions for the user of vehicle 1. For example, the control unit 21 may acquire visit history data of at least one of restaurants and tourist attractions for another person (i.e., a person different from the user of vehicle 1).

[0024] In the process of step S101 in Figure 2, the control unit 21 may select at least one of the following as features based on the visit history data of the user of vehicle 1: the genre of restaurants visited by the user (e.g., Japanese food, Western food, Chinese food, etc.) and the genre of tourist spots visited by the user (e.g., nature, leisure facilities, historical buildings, etc.). The control unit 21 may select as features the transition between locations that indicate the user visited one place and then another place, based on the visit history data of the user of vehicle 1. The control unit 21 may select at least one of the following as features based on the visit history data of other people: the number of people who visited one restaurant and the number of people who visited one tourist spot. The control unit 21 may select as features the transition between locations that indicate another person visited one place and then another place, based on the visit history data of other people. The control unit 21 may select at least one of the following as features: current time, season, weather, vehicle data, road information, business hours, site area, and parking information.

[0025] For example, if the visit history data of the user of vehicle 1 does not include any conditions that match or are similar to the current conditions of vehicle 1 (e.g., driving location, time of day, weather, etc.), the control unit 21 does not need to use the visit history data of the user of vehicle 1 in the processing of step S101.

[0026] In step S102 of Figure 2, the control unit 21 may obtain, for example, the formula “Recommendation score = x1 × α + x2 × β + ...” as the selection criterion. In the above formula, “x1, x2, ...” represent features, and “α, β, ...” represent parameters (for example, weighting parameters). The above formula is an example of a calculation formula for merging the features selected in step S101. In step S102, the control unit 21 may obtain, for example, exclusion filters such as “Exclude outdoor facilities when it is raining” or “Exclude facilities outside of business hours” as the selection criterion.

[0027] In step S105 of Figure 2, the control unit 21 controls the display unit 24 so that the recommendation results and reasons for recommendation are displayed. In this case, the display unit 24 may display an image such as that shown in Figure 3. In Figure 3, triangle C represents vehicle 1, and the solid line R represents the road on which vehicle 1 is traveling. The "Recommended Place" shown in the upper right of Figure 3 is an example of a recommendation result and reason for recommendation. In Figure 3, the numbers enclosed in squares around the solid line R correspond to the positions of "Top 1," "Top 2," and "Top 3" of the recommended places. "Popularity," "Distance," "Preference," and "Set Visit" included in the recommended places correspond to an example of a reason for recommendation. As shown in Figure 3, the control unit 21 may control the display unit 24 so that the recommendation results and reasons for recommendation are displayed on the image for route guidance. Note that the method of displaying recommendation results and reasons for recommendation as shown in Figure 3 may be called a push notification method.

[0028] Alternatively, in the process of step S105, the display unit 24 may display an image such as that shown in Figure 4. For example, if the operation shown in Figure 2 occurs because the user of vehicle 1 inputs "recommendations around Mt. Fuji" using the operation unit 25, the display unit 24 may display an image such as that shown in Figure 4. The method of displaying the recommendation results and reasons for recommendation as shown in Figure 4 may also be called a pull-type notification method. For example, if the user of vehicle 1 performs a location search using the operation unit 25, the control unit 21 may control the display unit 24 to display the recommendation results and reasons for recommendation using a pull-type notification method. For example, if the user of vehicle 1 has not set a destination in the navigation device 20, the control unit 21 may control the display unit 24 to display the recommendation results and reasons for recommendation using a push-type notification method.

[0029] In step S105, after the recommendation result and recommendation reason are displayed, the control unit 21 may control the display unit 24 to display the evaluation content. For example, if the user of vehicle 1 selects "Top 1: △△ store" as shown in Figure 3 or Figure 4, the display unit 24 may display an image for a questionnaire survey as shown in Figure 5. In Figure 5, "1. Is the timing and location of the recommendation information appropriate this time?", "2. Are you satisfied with the recommendation result this time?", and "3. Which factor was mainly considered?" are examples of evaluation content. Furthermore, even if the user of vehicle 1 does not operate the operation unit 25 within the second predetermined time elapsed after the recommendation result and recommendation reason are displayed (in other words, if the user does not react to the recommendation result), the control unit 21 may control the display unit 24 to display the evaluation content.

[0030] For example, if, after an image like the one shown in Figure 5 is displayed, the user of vehicle 1 answers a questionnaire using the operation unit 25, the control unit 21 may determine in step S106 of Figure 2 that there is user feedback. On the other hand, if the user does not answer the questionnaire, the control unit 21 may determine in step S106 that there is no user feedback.

[0031] An example of the process in step S107 in Figure 2 will be explained with reference to the flowchart in Figure 6. Here, an example of the process in step S107 when the selection criterion is a learning model will be described. In the process of step S107, the control unit 21 may improve the recommendation logic by performing reinforcement learning of the learning model using, for example, RLHF (Reinforcement Learning from Human Feedback).

[0032] In Figure 6, the control unit 21 processes user feedback for vehicle 1 (for example, responses to the questionnaire described in Figure 5) (step S201). For example, the control unit 21 may generate a policy function to adjust the parameters related to the features (for example, weighting parameters) based on the user feedback.

[0033] Next, the control unit 21 determines whether or not to improve the rules related to the recommendation logic (step S202). For example, the control unit 21 may determine the change in satisfaction level of the user of vehicle 1 from the satisfaction level based on the previous feedback and the satisfaction level based on the current feedback. If the change in satisfaction level is greater than a predetermined threshold, the control unit 21 may decide not to improve the rules. On the other hand, if the change in satisfaction level is less than a predetermined threshold, the control unit 21 may decide to improve the rules. If there is no previous feedback, the change in user satisfaction level may be determined from the initial value related to satisfaction and the satisfaction level based on the current feedback. Note that since the rules are improved when the change in satisfaction level is greater than a threshold, the condition that the change in satisfaction level is greater than a threshold may be called an improvement condition.

[0034] If it is determined in step S202 that no improvement to the rule is to be made (step S202: No), then, for example, the control unit 21 saves the policy function generated in step S201 and terminates the process shown in Figure 6 (in other words, the process in step S107 in Figure 2 is terminated).

[0035] If it is determined in step S202 that the rule should be improved (step S202: Yes), the control unit 21 modifies the rule based on the policy function generated in step S201 (step S203). The control unit 21 may determine the correspondence between changes in the rule parameters (for example, parameters related to features) and changes in user satisfaction. Based on the determined correspondence, the control unit 21 may modify the rule by estimating parameters such that the change in user satisfaction exceeds a predetermined threshold.

[0036] A specific example of the process in step S203 is described below. The learning model used to generate the recommendation results and recommendation reasons in the process in step S103 of Figure 2 is referred to as the initial learning model. The control unit 21 may generate a tuned learning model based on the policy function and the initial learning model. For example, the control unit 21 may generate a tuned learning model using PPO (Proximal Policy Optimization) or TRPO (Trust Region Policy Optimization).

[0037] For example, the control unit 21 may input the features selected in step S101 of Figure 2 to the initial learning model and the tuned learning model. The control unit 21 may obtain a reward value by inputting the output result of the tuned learning model into a predetermined reward function. The control unit 21 may obtain a penalty corresponding to the difference between the output result of the initial learning model and the output result of the tuned learning model. The control unit 21 may correct the reward value using the penalty. The control unit 21 may change the policy function generated in step S201 using the corrected reward value. The policy function may correspond to an example of a rule related to the recommendation logic. Improving the recommendation logic may mean changing the policy function.

[0038] After the processing in step S203, the control unit 21 stores the modified rule (e.g., the modified policy function described above) in the storage unit 22 (step S204). In parallel with the processing in step S204, the control unit 21 may control the display unit 24 so that the modified rule is presented to the user of vehicle 1 (step S205). In this case, the control unit 21 modifies the modified rule (e.g., the modified policy function described above) in a manner that the user can understand. The manner that the user can understand may include, for example, a graph showing the effect of stochastic gradient descent (e.g., how the value of the objective function converges). The manner that the user can understand may also include, for example, an explanatory text describing the modified rule. Note that the processing in step S205 may be performed only when requested by the user using the operation unit 25.

[0039] (Second specific example) As a second specific example of the operation of the navigation device 20, the operation of recommending rest stops will be explained with reference to Figure 2. Note that, in this second specific example, explanations that overlap with the explanation of the first specific example described above will be omitted as appropriate.

[0040] For example, the control unit 21 may acquire the user's driving history data for vehicle 1. The control unit 21 may acquire at least one of the data indicating the behavior of vehicle 1 and the data indicating the user's state in vehicle 1. In the process of step S101 in Figure 2, the control unit 21 may select the user's driving time and stopping timing as features from the acquired driving history data. The control unit 21 may select information indicating the user's driving state (e.g., the number of sudden brakings, the number of sudden accelerations, the number of yawns, etc.) as features from at least one of the data indicating the behavior of vehicle 1 and the data indicating the user's state in vehicle 1. The control unit 21 may select information related to the external environment of vehicle 1 (e.g., road information, weather, etc.) as features. The control unit 21 may select attribute data related to places where the user of vehicle 1 can rest (e.g., whether there is a toilet, whether there is a parking lot, etc.) as features. The control unit 21 may select at least one of the destination set in the navigation device 20, the time required to reach the destination, and the position of vehicle 1 at a certain time as features.

[0041] In step S102 of Figure 2, the control unit 21 may obtain, for example, a formula such as “Recommendation score = x1 × α + x2 × β + ...” as a selection criterion. In step S102, the control unit 21 may obtain, for example, an exclusion filter such as “Exclude facilities unsuitable for short breaks” or “Exclude facilities unsuitable for long breaks” as a selection criterion.

[0042] For example, the control unit 21 may estimate the timing of the user's rest based on at least one of the following: the user's driving history data for vehicle 1, the destination set in the navigation device 20, road information (e.g., traffic congestion information, construction information, etc.), and the location of a suitable rest stop. The control unit 21 may also include the estimated rest timing in the selection criteria.

[0043] The recommendation reasons generated in step S103 of Figure 2 may include at least one of the following: "close distance," "toilets available," "food available," "smoking allowed," "napping allowed," and "overnight stay available."

[0044] (Third specific example) As a third specific example of the operation of the navigation device 20, the operation of recommending a stopover location where vehicle maintenance can be performed will be explained with reference to Figure 2. Note that for this third specific example, explanations that overlap with the explanation of the first specific example described above will be omitted as appropriate.

[0045] In step S101 of Figure 2, the control unit 21 may select data related to the parts of the vehicle 1 as features. The data related to the parts is not limited to data that directly indicates the state of the parts, but may also be data that indirectly indicates the state of the parts. For example, the state of the engine may be indicated as the temperature of the coolant that cools the engine. Data related to consumable parts may be represented by the mileage of the vehicle 1, or by the elapsed time since the last replacement.

[0046] The control unit 21 may acquire at least one of the vehicle inspection history data, car wash history data, and maintenance history data of vehicle 1. Based on the vehicle inspection history data, the control unit 21 may select the date and time of the vehicle inspection and the location of the vehicle inspection as features. Based on the car wash history data, the control unit 21 may select the date and time of the car wash and the location of the car wash as features. Based on the maintenance history data, the control unit 21 may select the date and time of the maintenance and the location of the maintenance as features.

[0047] In step S102 of Figure 2, the control unit 21 may obtain, for example, a formula such as “Recommendation score = x1 × α + x2 × β + ...” as a selection criterion. In step S102, the control unit 21 may obtain, for example, an exclusion filter such as “Exclude facilities unsuitable for vehicle maintenance” as a selection criterion.

[0048] In parallel with the processing in steps S101 and S102 of Figure 2, the control unit 21 may perform abnormality detection on the parts of the vehicle 1. For example, the control unit 21 may perform abnormality detection on the parts of the vehicle 1 using a One Class Support Vector Machine. For example, the control unit 21 may classify the degree of abnormality, which indicates the severity of the abnormality, into "high," "medium," or "low" according to the number of abnormality data.

[0049] The recommendation reasons generated in step S103 of Figure 2 may include at least one of the following: "close distance," "frequently used," and "highly rated." In step S105 of Figure 2, the control unit 21 may control the display unit 24 so that the display mode of the recommendation results changes according to the degree of abnormality. For example, if the degree of abnormality is "high," the control unit 21 may control the display unit 24 so that a stopover location where vehicle maintenance can be performed is displayed along with the message "Emergency response requested." For example, if the degree of abnormality is "medium," the control unit 21 may control the display unit 24 so that a stopover location where vehicle maintenance can be performed is displayed along with the message "Warning information."

[0050] (Technical effects) In the navigation system 20, the rules related to the recommendation logic are improved when feedback is received from the user of vehicle 1. Therefore, the method of recommending places of interest by the navigation system 20 can improve the rules related to the recommendation logic. As a result, more appropriate places of interest can be recommended.

[0051] The embodiments of the invention derived from the above-described embodiments are described below.

[0052] A recommendation method according to one aspect of the invention is a recommendation method for recommending a stopover location to a user based on predetermined rules, comprising: a receiving step of receiving input of feedback information from the user indicating whether the recommendation of the stopover location was appropriate or not; a determination step of determining whether the feedback information satisfies improvement conditions related to the predetermined rules; and an update step of updating the predetermined rules based on the feedback information if it is determined that the feedback information satisfies the improvement conditions.

[0053] In the above-described embodiment, for example, the display of an image for a questionnaire survey shown in Figure 5 corresponds to an example of the reception process, the processing in step S202 in Figure 6 corresponds to an example of the determination process, and the processing in step S203 in Figure 6 corresponds to an example of the update process.

[0054] The recommendation method may include a rule presentation step in which the updated rules from the update step are presented to the user. In the embodiment described above, the process in step S205 of Figure 6 corresponds to an example of the presentation step.

[0055] The recommendation method may include a reason presentation step in which the user is presented with the reasons for the recommendation of the stopover locations. In the embodiment described above, the process in step S105 of Figure 2 corresponds to an example of the reason presentation step.

[0056] In this recommendation method, the user is in the vehicle, and the recommended stops may be presented to the user at a time that does not affect the vehicle's movement.

[0057] The present invention is not limited to the embodiments described above, and can be modified as appropriate without contradicting the gist or idea of ​​the invention as can be read from the claims and specification as a whole. Such modifications to the recommendation method are also included within the technical scope of the present invention. [Explanation of Symbols]

[0058] 1...Vehicle, 20...Navigation system, 21...Control unit, 22...Storage unit, 23...In-vehicle communication interface, 24...Display unit, 25...Operation unit

Claims

1. A recommendation method for a navigation system installed in a vehicle, which recommends a stopover to a user riding in the vehicle based on predetermined rules, A receiving step in which, after a second image different from the first image is displayed on the navigation device in place of the first image which includes one or more recommended stops, the navigation device receives feedback information from the user indicating whether or not the recommendations for stops were appropriate. The navigation device includes a determination step of determining whether the feedback information satisfies the improvement conditions related to the predetermined rule, If the feedback information is determined to satisfy the improvement conditions, the navigation device performs an update step in which it updates the predetermined rule based on the feedback information. Recommendation methods including

2. The navigation device includes a rule presentation step of presenting the updated rules in the update step to the user. The recommendation method described in claim 1.

3. The navigation device includes a reason presentation step of presenting the user with the reasons for recommending a place to stop. The recommendation method described in claim 1.

4. The navigation device presents recommended stops to the user at a time that does not affect the vehicle's movement. The recommendation method described in claim 1.

Images