Intelligent management method and system for tourism service supply chain

Abstract

The invention relates to the technical field of tourism intelligent management, in particular to a tourism service supply chain intelligent management method and system. When a new travel link is about to be started, the tour guide terminal sends the new travel link to be started to the corresponding user terminal; the user terminal generates feedback evaluation information determined by the user, wherein the feedback evaluation information can reflect the subjective impression of the tourist on the tourism link; when the feedback evaluation information of the same evaluation link reaches a preset quantity, marking a supplier corresponding to the tourism link as a target supplier; the management terminal determines a user score value corresponding to the target supplier based on the feedback evaluation information corresponding to the target supplier; and when the user score value is lower than a first preset value, marking the target supplier as a to-be-improved supplier, generating supervision alarm information, and sending the supervision alarm information to a supplier terminal corresponding to the to-be-improved supplier so as to supervise the to-be-improved supplier.

Description

Technical Field

[0001] This invention relates to the field of smart tourism management technology, specifically to a smart management method and system for the tourism service supply chain. Background Technology

[0002] As the national economic level rises and people's consumption power increases, the tourism industry is booming. A complete tourism product cannot be separated from the joint assistance of suppliers at all stages (such as transportation, food and accommodation suppliers). However, current tourism suppliers cannot accurately understand tourists' needs, and their business logic is still completed through traditional manual services such as telephone. During the tourism process, tourism platforms are also unable to effectively supervise the services or products of various suppliers in a timely manner. Summary of the Invention

[0003] The main objective of this invention is to provide a smart management method and system for the tourism service supply chain, which aims to solve the problem that tourism platforms are unable to effectively monitor the services or products of various suppliers in a timely manner during the tourism process.

[0004] The technical solution proposed in this invention is as follows: A smart management method for the tourism service supply chain is applied to a smart management system for the tourism service supply chain; the system includes a management terminal, a user terminal, and a tour guide terminal; the user terminal and the tour guide terminal are both communicatively connected to the management terminal; the method includes: Before the start of the tour, the management terminal will establish a correspondence between the tour guide terminal and the user terminal participating in the tour. The tour includes multiple tour components. Whenever a new tour segment is about to begin, the tour guide terminal sends the information about the upcoming tour segment to the corresponding user terminal. Each tour segment has a unique supplier. The user terminal generates feedback evaluation information corresponding to the completed tourism segments in this tour itinerary input by the user, and sends it to the management terminal. The feedback information includes evaluation type, evaluation text, and evaluation segment. The evaluation type can be any one of positive, negative, or neutral. The evaluation segment is any one of the tourism segments received from the tour guide terminal. When the number of feedback evaluation messages received by the management terminal for the same evaluation stage reaches a preset number, the supplier corresponding to the same evaluation stage is marked as the target supplier. The management terminal determines the user rating value of the target supplier based on the feedback and evaluation information of the target supplier. The higher the user rating value, the higher the tourist's satisfaction with the target supplier's service or product. The management terminal marks the target supplier whose user rating is lower than the first preset value as a supplier to be improved, generates a monitoring alarm, and sends the monitoring alarm to the supplier terminal corresponding to the supplier to be improved.

[0005] Preferred options also include: The management terminal constructs and trains a multinomial Naive Bayes model. The management terminal determines the user rating value for the target supplier based on the feedback and evaluation information corresponding to the target supplier, including: The management terminal marks the feedback evaluation information corresponding to the target supplier as target information; The management terminal performs sentiment assessments on the evaluation texts in each target information sequentially based on a trained multinomial Naive Bayes model to determine the sentiment type of the evaluation texts, wherein the sentiment type is any one of positive, negative, or neutral. The management terminal marks target information whose emotion type and evaluation type are consistent as normal information; The management terminal calculates the user rating value corresponding to the target supplier based on the target information and normal information.

[0006] Preferably, the formula for calculating the user rating value corresponding to the target supplier based on the target information and normal information by the management terminal is as follows: , In the formula, The user rating value corresponding to the target supplier; The number of normal information items in the target information corresponding to the target supplier; This represents the number of positive feedback ratings in the normal information.

[0007] Preferably, the management terminal constructs and trains a multinomial Naive Bayes model, including: The management terminal constructs a multinomial Naive Bayes model. The management terminal acquires a first text set and a second text set, wherein the first text set includes multiple first texts with a positive sentiment type, and the second text set includes multiple second texts with a negative sentiment type. The management terminal performs word segmentation on the first text set and the second text set to obtain multiple training words, and trains a multinomial Naive Bayes model based on the first text set, the second text set and each training word. The management terminal performs sentiment assessments on the evaluation texts in each target information sequentially based on a trained multinomial Naive Bayes model to determine the sentiment type of the evaluation texts, including: The management terminal sequentially segments the evaluation text in each target information to obtain the evaluation words corresponding to the evaluation text; The management terminal calculates the probability that the sentiment type of the evaluation text is positive and the probability that the sentiment type of the evaluation text is negative, based on the trained multinomial Naive Bayes model and each evaluation word in the evaluation text. When the probability of a text being positive is greater than the probability of it being negative, the management terminal determines the sentiment of the text to be positive. When the probability of a text being positive is less than the probability of a text being negative, the management terminal determines the sentiment of the text to be negative. When the probability that the sentiment type of the evaluation text is positive is equal to the probability that the sentiment type of the evaluation text is negative, the management terminal determines the sentiment type of the evaluation text as neutral.

[0008] Preferably, the feedback evaluation information further includes a feedback image captured by the user through the camera module of the user terminal; the system further includes a cloud server communicatively connected to the management terminal; the management terminal further includes a display screen and a camera; the user terminal generates the feedback evaluation information determined by the user and sends it to the management terminal, and then further includes: The management terminal generates a positioning page on the display screen. The positioning page has prompt text in the center and a positioning icon block in each of the four corners of the display screen. The positioning page is a rectangular page. The prompt text is used to prompt the customer to focus on each positioning icon block in turn. The management terminal acquires real-time facial video of the management personnel while the display screen shows the location page via a camera, and marks it as the first target video; The management terminal displays the feedback image on a screen and acquires real-time facial video of the management personnel while the feedback image is displayed on the screen via a camera, which is marked as the second target video. The feedback image is a rectangular image, and the size of the feedback image displayed on the screen is the same as the size of the positioning page. The management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, and the feedback image; The management terminal reduces the resolution of areas other than key areas in the feedback image to generate a simplified image; The management terminal sends the simplified image to the cloud server for storage.

[0009] Preferably, the management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, and the feedback image, including: The management terminal performs image frame decomposition on the first target video to obtain multiple frame images. The obtained frame images are then mirrored and marked as the first frame image, wherein the first frame image is a rectangular image including the customer's face. The management terminal performs image recognition on each first frame image to determine the first eyeball marker point of each first frame image. The first eyeball marker point is the center point of the line segment connecting the center point of the left eyeball and the center point of the right eyeball in the first frame image. The management terminal uses the first eyeball markers closest to the four corners of the first frame image among multiple first frame images as the first target point, the second target point, the third target point, and the fourth target point, respectively. The management terminal takes the distance between the first target point and the third target point as the first distance value, and the distance between the second target point and the fourth target point as the second distance value. It takes the average of the first distance value and the second distance value and marks it as the diagonal average, where the first target point and the third target point are diagonal points, and the second target point and the fourth target point are diagonal points. The management terminal obtains the length value of any diagonal of the feedback image and marks it as the diagonal length value; it also obtains the ratio of the diagonal average value to the diagonal length value and marks it as the ratio. The management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, the feedback image, and the ratio.

[0010] Preferably, the management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, the feedback image, and the ratio, including: The management terminal performs image frame decomposition on the second target video to obtain multiple frame images. The obtained frame images are then mirrored and marked as the second frame image. The second frame image is a rectangular image including the customer's face, and the size of the second frame image is the same as that of the first frame image. The management terminal performs image recognition on each second frame image to determine the second eyeball marker point of each second frame image. The second eyeball marker point is the center point of the line segment connecting the center point of the left eyeball and the center point of the right eyeball in the second frame image. The management terminal constructs a first polar coordinate system based on multiple first frame images. The pole of the first polar coordinate system is the intersection of the line segment connecting the first target point and the third target point with the line segment connecting the second target point and the fourth target point. The polar axis of the first polar coordinate system is a ray extending horizontally to the right. The management terminal constructs a second polar coordinate system based on the feedback image, wherein the pole of the second polar coordinate system is the intersection of the two diagonals of the feedback image, and the polar axis of the second polar coordinate system is a ray extending horizontally to the right. The management terminal marks the second eyeball marker point in the second frame image that remains in the same position for a duration longer than a preset duration as the target eyeball marker point; The management terminal substitutes the first polar coordinate system into the second frame image to determine the polar coordinates of the target eyeball marker in the second frame image in the first polar coordinate system. ; The management terminal is based on a ratio and Calculate the polar coordinates of the focal point corresponding to the second eye marker in the feedback image in the second polar coordinate system. : , In the formula, It is a ratio; The management terminal determines the key area based on the polar coordinates of the focal point. The key area is a circular area with the focal point as the center and a third preset value as the radius. When there are multiple key areas and they intersect, the intersecting area is taken only once. When a part of the key area exceeds the feedback image, the part that exceeds the feedback image is discarded.

[0011] Preferably, the management terminal reduces the resolution of areas other than key areas of the feedback image to generate a simplified image, including: The management terminal determines the buffer area based on the polar coordinates of the focal point. The buffer area is a circular area with the focal point as the center and the fourth preset value as the radius. When there are multiple buffer areas and there are intersecting areas, the intersecting area is only taken once. When part of the buffer area exceeds the feedback image, the part that exceeds the feedback image is discarded. The fourth preset value is greater than the third preset value. The management terminal marks the area outside the key area in the buffer region of the feedback image as the middle area, and marks the area outside the buffer region in the feedback image as the non-key area; The management terminal reduces the resolution of both non-critical areas and intermediate areas to generate a simplified image, wherein the resolution of non-critical areas is lower than that of intermediate areas.

[0012] This invention also proposes a smart management system for the tourism service supply chain. The system, which applies a smart management method for the tourism service supply chain, includes a management terminal, a user terminal, and a tour guide terminal; the user terminal and the tour guide terminal are both communicatively connected to the management terminal.

[0013] The above technical solution can achieve the following beneficial effects: The intelligent management method for the tourism service supply chain proposed in this invention can solve the problem that tourism platforms cannot effectively supervise the services or products of various suppliers in a timely manner during the tourism process. Firstly, before the start of the tour, the management terminal establishes a correspondence between the tour guide terminals and user terminals participating in the tour. Then, whenever a new tour segment is about to begin, the tour guide terminal sends the upcoming segment to the corresponding user terminal. The user terminal then generates feedback evaluation information corresponding to the completed tour segments input by the user. This feedback evaluation information reflects the tourist's subjective impression of a completed segment. When the feedback evaluation information for the same segment reaches a preset number, it indicates that a significant issue has arisen in that segment. When multiple tourists provide feedback simultaneously, indicating a high level of interest in the tourism environment, the supplier corresponding to that tourism segment is marked as a target supplier. The management terminal then determines the user rating for the target supplier based on the feedback. A higher user rating indicates greater tourist satisfaction with the target supplier's services or products. When a user rating falls below a preset threshold, it indicates that the supplier's services or products are not satisfactory to the tourist and require improvement. Therefore, the target supplier is marked as a supplier requiring improvement, and a monitoring alarm is generated and sent to the supplier's terminal to monitor the supplier and improve the tourist's user experience. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating the steps of a first embodiment of a smart management method for the tourism service supply chain proposed in this invention. Figure 2 This is a schematic diagram of the positioning page in the sixth embodiment of the intelligent management method for the tourism service supply chain proposed in this invention. Detailed Implementation

[0016] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0017] This invention proposes a smart management method and system for the tourism service supply chain.

[0018] As attached Figure 1As shown, in the first embodiment of the intelligent management method for the tourism service supply chain proposed in this invention, the method is applied to an intelligent management system for the tourism service supply chain; the system includes a management terminal (an intelligent terminal used by management personnel, such as a computer), a user terminal (an intelligent terminal carried by tourists at all times, such as a smartphone), and a tour guide terminal (an intelligent terminal carried by tour guides at all times, such as a smartphone); the user terminal and the tour guide terminal are both communicatively connected to the management terminal; this embodiment includes the following steps: Step S110: Before the start of the tour, the management terminal establishes a correspondence between the tour guide terminal and the user terminal participating in this tour. The tour includes multiple tour components.

[0019] Step S120: Whenever a new tour segment is about to begin, the tour guide terminal sends the upcoming new tour segment to the corresponding user terminal, wherein each tour segment corresponds to a unique supplier.

[0020] Specifically, a tour itinerary involves multiple independent travel segments, such as transportation, accommodation, dining, or visiting scenic spots. Each segment is served by a corresponding supplier. For example, catering supplier A is responsible for providing food services for a certain catering segment. In the application scenario of this application, only one supplier is set up for each travel segment, so that when users evaluate a certain travel segment, the management terminal can directly and accurately find the supplier corresponding to that travel segment.

[0021] Step S130: The user terminal generates feedback evaluation information corresponding to the completed tourism segments in this tour itinerary input by the user, and sends it to the management terminal. The feedback information includes evaluation type, evaluation text, and evaluation segment. The evaluation type can be any one of positive, negative, or neutral. The evaluation segment is any one of the tourism segments received from the tour guide terminal (that is, the tourism segment evaluated by the user, for example, if the user evaluates a certain catering segment, then the evaluation segment of the feedback evaluation information is the catering segment).

[0022] Specifically, the feedback and evaluation information here is generated by tourists through their own input on their user terminals, and can reflect the tourists' subjective impressions of this tour.

[0023] Step S140: When the number of feedback evaluation information received by the management terminal for the same evaluation stage reaches a preset number (the preset number is determined by a certain proportion of the total number of tourists in this tour, such as 30%), the supplier corresponding to the same evaluation stage is marked as the target supplier.

[0024] Specifically, when the number of feedback comments for the same evaluation stage reaches a preset number, it proves that a large number of tourists are simultaneously evaluating this tourism stage, indicating that the tourism environment has received high attention. Therefore, the supplier corresponding to this tourism stage is marked as a target supplier, so that the target supplier can be evaluated and analyzed in the future to determine whether the target supplier's service meets the standards.

[0025] Step S150: The management terminal determines the user rating value of the target supplier based on the feedback evaluation information of the target supplier. The higher the user rating value, the higher the tourist's satisfaction with the target supplier's service or product.

[0026] Specifically, the user rating calculated here is greater than 0 and less than or equal to 1.

[0027] Step S160: The management terminal marks the target supplier whose user rating value is lower than the first preset value (e.g., 0.3) as a supplier to be improved, generates a monitoring alarm message, and sends the monitoring alarm message to the supplier terminal corresponding to the supplier to be improved.

[0028] Specifically, when a user rating is lower than the first preset value, it means that the service or product provided by the supplier corresponding to that user rating is not satisfactory to the tourist and needs to be improved. Therefore, the target supplier is marked as a supplier to be improved, and a monitoring alarm is generated and sent to the supplier terminal corresponding to the supplier to be improved in order to monitor the supplier to be improved and improve the tourist's user experience.

[0029] The intelligent management method for the tourism service supply chain proposed in this invention can solve the problem that tourism platforms cannot effectively supervise the services or products of various suppliers in a timely manner during the tourism process. Firstly, before the start of the tour, the management terminal establishes a correspondence between the tour guide terminals and user terminals participating in the tour. Then, whenever a new tour segment is about to begin, the tour guide terminal sends the upcoming segment to the corresponding user terminal. The user terminal then generates feedback evaluation information corresponding to the completed tour segments input by the user. This feedback evaluation information reflects the tourist's subjective impression of a completed segment. When the feedback evaluation information for the same segment reaches a preset number, it indicates that a significant issue has arisen in that segment. When multiple tourists provide feedback simultaneously, indicating a high level of interest in the tourism environment, the supplier corresponding to that tourism segment is marked as a target supplier. The management terminal then determines the user rating for the target supplier based on the feedback. A higher user rating indicates greater tourist satisfaction with the target supplier's services or products. When a user rating falls below a preset threshold, it indicates that the supplier's services or products are not satisfactory to the tourist and require improvement. Therefore, the target supplier is marked as a supplier requiring improvement, and a monitoring alarm is generated and sent to the supplier's terminal to monitor the supplier and improve the tourist's user experience.

[0030] In a second embodiment of the intelligent management method for the tourism service supply chain proposed in this invention, based on the first embodiment, this embodiment further includes the following steps: Step S210: The management terminal constructs and trains a multinomial Naive Bayes model.

[0031] Specifically, multinomial Naive Bayes is a special case of the Naive Bayes family. It is based on Bayes' theorem and makes a "naive" independence assumption. It is particularly suitable for count data with discrete features, especially text data, and is especially suitable for determining whether the sentiment of the text is positive or negative.

[0032] Step S140 includes the following steps: Step S220: The management terminal marks the feedback evaluation information corresponding to the target supplier as target information.

[0033] Step S230: The management terminal performs sentiment assessment on the evaluation text in each target information sequentially based on the trained multinomial Naive Bayes model to determine the sentiment type of the evaluation text, wherein the sentiment type is any one of positive, negative and neutral.

[0034] Step S240: The management terminal marks target information whose emotion type and evaluation type are consistent as normal information.

[0035] Specifically, the sentiment type of the target information is evaluated using a trained multinomial Naive Bayes model, while the rating type is selected and input by the user. Normally, the sentiment type and rating type of the target information are consistent, and target information with consistent sentiment and rating types is marked as normal information. However, it is possible that the user may make a mistake in inputting the rating text and rating type, resulting in a contradiction, i.e., the sentiment type and rating type are inconsistent. In this case, the target information cannot be used for subsequent calculation of the user rating value corresponding to the target supplier, and therefore needs to be discarded.

[0036] Step S250: The management terminal calculates the user rating value corresponding to the target supplier based on the target information and normal information.

[0037] In the third embodiment of the intelligent management method for tourism service supply chain proposed in this invention, based on the second embodiment, the management terminal includes a display screen; step S240, followed by the following steps: Step S310: The management terminal marks target information other than normal information as abnormal information.

[0038] Specifically, the abnormal information here refers to the target information where the sentiment type and evaluation type are inconsistent. The evaluation type and evaluation text are contradictory, so the target information needs to be removed. Therefore, the target information is marked as abnormal information.

[0039] Step S320: The management terminal obtains the ratio of the number of abnormal information to the number of target information and marks it as the abnormal ratio.

[0040] Step S330: The management terminal determines whether the abnormal ratio is greater than a second preset value (e.g., 0.2).

[0041] If so, proceed to step S340: The management terminal displays the target information sequentially on the display screen, and then obtains the correction score value corresponding to the target supplier input by the management personnel.

[0042] Specifically, under normal circumstances, the number of abnormal information should be relatively small, so the anomaly ratio will be relatively small; if the anomaly ratio is greater than the second preset value, it proves that there is a lot of abnormal information in the feedback evaluation information corresponding to this tourism link; manual screening is required to more accurately determine the corrective score value corresponding to the target supplier.

[0043] If not, proceed to step S250.

[0044] Specifically, if not, it means that there is indeed very little abnormal information, which is consistent with normal logic, and the subsequent step S250 can be continued.

[0045] In the fourth embodiment of the intelligent management method for tourism service supply chain proposed in this invention, based on the second embodiment, the calculation formula for the user rating value corresponding to the target supplier by the management terminal based on target information and normal information is as follows: , In the formula, The user rating value corresponding to the target supplier; The number of normal information items in the target information corresponding to the target supplier; This represents the number of positive feedback ratings in the normal information.

[0046] Specifically, this embodiment provides a formula for calculating the user rating value corresponding to the target supplier. It can be seen that the user rating value is greater than 0 and less than or equal to 1. The higher the user rating value, the higher the tourist's satisfaction with the target supplier's service or product.

[0047] In the fifth embodiment of the intelligent management method for tourism service supply chain proposed in this invention, based on the second embodiment, step S210 includes the following steps: Step S510: The management terminal constructs a multinomial Naive Bayes model.

[0048] Step S520: The management terminal acquires a first text set and a second text set, wherein the first text set includes multiple first texts with a positive sentiment type, and the second text set includes multiple second texts with a negative sentiment type.

[0049] Specifically, for example, the first text set is: "The scenery at the scenic spot is beautiful, and the service is excellent. I am very satisfied." "The tour guide was professional and enthusiastic, the itinerary was well-planned, and the value for money was excellent." "The hotel is clean and comfortable, with a great location. Highly recommended." "This trip was a great experience, I'll definitely come back next time." "A very cost-effective travel product, I had a lot of fun." "Convenient transportation, beautiful environment, suitable for vacation" "The staff were very friendly and resolved the issues promptly." "The facilities are excellent, making it suitable for family trips, and the children are very happy." For example, the second text set is: "Too many people, long queues, terrible experience." "The hotel's hygiene is poor, and the room smells bad." "The price is too high and does not meet expectations at all." "The service was terrible, I'll never come back." "The facilities are outdated, and many projects are under repair." "The tour guide was unprofessional, and the itinerary was disorganized." "The food was unpalatable and there were very few choices." "The transportation is inconvenient, and there's not much to do around here." Step S530: The management terminal performs word segmentation on the first text set and the second text set to obtain multiple training words, and trains a multinomial Naive Bayes model based on the first text set, the second text set, and each training word, specifically including the following steps: Step S531: The management terminal calculates the positive prior probability. and negative prior probability : , , In the formula, The number of texts in the first text; The number of the second text; Step S532: For each training word, the management terminal calculates the conditional probability. and : , In this formula, the nth training word W is used as an example for calculation, and other training words are calculated in the same way. The total number of occurrences of the word W (e.g., "convenience") in the first text set is used for training. This represents the total number of occurrences of all training words in the first text set. The total number of training words; , In this formula, the nth training word W is used as an example for calculation, and other training words are calculated in the same way. The total number of occurrences of the word W (e.g., "convenience") in the second text set is trained; This represents the total number of occurrences of all training words in the second text set.

[0050] Step S230 includes the following steps: Step S540: The management terminal sequentially performs word segmentation on the evaluation text in each target information to obtain the evaluation words corresponding to the evaluation text.

[0051] Step S550: The management terminal calculates the probability that the sentiment type of the evaluation text is positive and the probability that the sentiment type of the evaluation text is negative, based on the completed multinomial Naive Bayes model and each evaluation word of the evaluation text.

[0052] The specific calculation formula is as follows: Π_{n=1}^{N} , Π_{n=1}^{N} , In the formula, To evaluate the probability that the sentiment type of the text is positive; Π represents the probability that the sentiment type of the evaluation text is negative; n represents the nth evaluation word; and N is the total number of evaluation words obtained after word segmentation of the evaluation text.

[0053] Step S560: When the probability that the sentiment type of the evaluation text is positive is greater than the probability that the sentiment type of the evaluation text is negative, the management terminal determines the sentiment type of the evaluation text as positive.

[0054] Step S570: When the probability that the sentiment type of the evaluation text is positive is less than the probability that the sentiment type of the evaluation text is negative, the management terminal determines the sentiment type of the evaluation text as negative.

[0055] Step S580: When the probability that the sentiment type of the evaluation text is positive is equal to the probability that the sentiment type of the evaluation text is negative, the management terminal determines the sentiment type of the evaluation text as neutral.

[0056] Specifically, this embodiment provides the steps for performing sentiment assessments on the evaluation texts in each target information sequentially based on a trained multinomial Naive Bayes model to determine the sentiment type of the evaluation texts.

[0057] In the sixth embodiment of the intelligent management method for tourism service supply chain proposed in this invention, based on the first embodiment, the feedback evaluation information further includes feedback images taken by the user through the camera module of the user terminal (i.e., images attached by the user when submitting feedback evaluation information, such as food images taken in the catering process); the system further includes a cloud server communicatively connected to the management terminal; the management terminal further includes a display screen and a camera; step S130, followed by the following steps: Step S610: The management terminal generates a positioning page on the display screen. The positioning page has prompt text in the center and one positioning icon block in each of the four corners of the display screen. The positioning page is a rectangular page. The prompt text is used to prompt the customer to focus on each positioning icon block in turn.

[0058] Specifically, the location page is shown in the attached image. Figure 2 As shown, by setting a positioning page, the eye position of the administrator can be located in advance. The eye position can reflect the current position of the administrator looking at the display screen. This allows the administrator to determine the focus point of the feedback image displayed on the screen by recognizing the eye position. This helps to identify which areas in the feedback image are key areas, so that the resolution of non-key areas can be reduced, thereby simplifying the feedback image, reducing the space occupied by the feedback image, and saving storage costs on the cloud server.

[0059] Step S620: The management terminal acquires real-time facial video of the management personnel while the display screen shows the positioning page through the camera, and marks it as the first target video.

[0060] Step S630: The management terminal displays the feedback image on the display screen and acquires real-time facial video of the management personnel while the feedback image is displayed on the display screen via a camera, and marks it as the second target video. The feedback image is a rectangular image, and the size of the feedback image displayed on the display screen is the same as the size of the positioning page.

[0061] Specifically, the second target video can reflect the level of attention that managers pay to various locations in the feedback image when viewing it, so as to help determine the key areas of the feedback image based on the first target video, the second target video, and the feedback image.

[0062] Step S640: The management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, and the feedback image.

[0063] Specifically, the key areas here are the areas that managers focus on when viewing the feedback image, proving that the image content in this area is the core content of the entire feedback image. The image content in non-key areas can be reduced in resolution to convert the feedback image into a smaller, simplified image. This reduces the space occupied by the feedback image when stored on the cloud server, while ensuring that the core content of the feedback image is still readable and understandable.

[0064] Step S650: The management terminal reduces the resolution of areas other than key areas in the feedback image to generate a simplified image.

[0065] Step S660: The management terminal sends the simplified image to the cloud server for storage.

[0066] In the seventh embodiment of the intelligent management method for tourism service supply chain proposed in this invention, based on the sixth embodiment, step S640 includes the following steps: Step S710: The management terminal performs image frame decomposition on the first target video to obtain multiple frame images. The obtained frame images are mirrored and flipped and marked as the first frame image, wherein the first frame image is a rectangular image including the customer's face.

[0067] Specifically, since the manager is facing the positioning page, the first target video captured by the camera needs to be mirrored after image frame decomposition to obtain the first frame image that can correctly reflect the relationship between the eye's focus point and the positioning page. For example, when the manager is staring at the upper left corner of the positioning page, the manager's eye is actually slightly to the upper right in the first target video captured by the camera. Therefore, mirroring is required to facilitate subsequent focus position correlation analysis.

[0068] Step S720: The management terminal performs image recognition on each first frame image to determine the first eyeball marker point of each first frame image, wherein the first eyeball marker point is the center point of the line segment connecting the center point of the left eyeball and the center point of the right eyeball in the first frame image.

[0069] Specifically, the first eye marker here can reflect the focus of the manager on the display screen. Therefore, by analyzing the position of the first eye marker, the manager's current viewing position on the display screen can be determined.

[0070] Step S730: The management terminal uses the first eyeball markers closest to the four corners of the first frame image among the multiple first frame images as the first target point, the second target point, the third target point, and the fourth target point, respectively.

[0071] Specifically, the management terminal designates the first eye-tracking marker closest to the top left corner of the first frame as the first target point (corresponding to the situation when the manager is looking at the positioning icon in the top left corner), the second eye-tracking marker closest to the top right corner of the first frame as the second target point (corresponding to the situation when the manager is looking at the positioning icon in the top right corner), the third eye-tracking marker closest to the bottom right corner of the first frame as the third target point (corresponding to the situation when the manager is looking at the positioning icon in the bottom right corner), and the fourth eye-tracking marker closest to the bottom left corner of the first frame as the fourth target point (corresponding to the situation when the manager is looking at the positioning icon in the bottom left corner). The area enclosed by these first, second, third, and fourth target points is the range of motion of the manager's gaze.

[0072] Step S740: The management terminal takes the distance between the first target point and the third target point as the first distance value (in pixels), and the distance between the second target point and the fourth target point as the second distance value (in pixels). It takes the average of the first distance value and the second distance value and marks it as the diagonal average value, where the first target point and the third target point are diagonal points, and the second target point and the fourth target point are diagonal points.

[0073] Specifically, since the first and second distance values ​​are relatively close, the average of the two (i.e., the diagonal average) is taken. This diagonal average can reflect the overall size of the area enclosed by the first, second, third, and fourth target points.

[0074] Step S750: The management terminal obtains the length value of any diagonal of the feedback image (in pixels), marks it as the diagonal length value, obtains the ratio of the diagonal average value to the diagonal length value, and marks it as the ratio.

[0075] Specifically, since the feedback image is rectangular, the length of any diagonal of the feedback image (i.e., the diagonal length value) can reflect the overall size of the feedback image; while the ratio of the average diagonal value to the diagonal length value (the ratio) can reflect the proportional relationship between the size of the manager's eye movement area and the size of the feedback image.

[0076] Step S760: The management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, the feedback image, and the ratio.

[0077] Specifically, once the ratio reflecting the size of the manager's eye movement area and the feedback image is known, the manager's eye movement trajectory in the second target video can be analyzed to determine the manager's eye focus position when viewing the feedback image, thereby identifying the key areas that the manager focuses on in the feedback image.

[0078] In the eighth embodiment of the intelligent management method for tourism service supply chain proposed in this invention, based on the seventh embodiment, step S760 includes the following steps: Step S810: The management terminal performs image frame decomposition on the second target video to obtain multiple frame images. The obtained frame images are mirrored and marked as the second frame image. The second frame image is a rectangular image including the customer's face, and the size of the second frame image is the same as that of the first frame image.

[0079] Specifically, since the manager is looking at the feedback image, the second target video captured by the camera needs to be mirrored after image frame decomposition to obtain a second frame image that can correctly reflect the correlation between the eye's focal point and the feedback image. For example, when the manager is staring at the upper left corner of the feedback image, the manager's eye is actually slightly to the upper right in the second target video captured by the camera. Therefore, mirroring is required to facilitate subsequent focus position correlation analysis.

[0080] Step S820: The management terminal performs image recognition on each second frame image to determine the second eyeball marker point of each second frame image, wherein the second eyeball marker point is the center point of the line segment connecting the center point of the left eyeball and the center point of the right eyeball in the second frame image.

[0081] Specifically, the second eye markers here can reflect the focus of the manager when viewing the feedback image on the display screen. Therefore, by analyzing the position of the second eye markers, the key areas of the feedback image can be inferred.

[0082] Step S830: The management terminal constructs a first polar coordinate system based on multiple first frame images, wherein the pole of the first polar coordinate system is the intersection of the line segment connecting the first target point and the third target point with the line segment connecting the second target point and the fourth target point, and the polar axis of the first polar coordinate system is a ray extending horizontally to the right.

[0083] Specifically, by establishing a first polar coordinate system, the position of the eye markers of managers can be analyzed and calculated in a concrete and quantitative manner.

[0084] Step S840: The management terminal constructs a second polar coordinate system based on the feedback image, wherein the pole of the second polar coordinate system is the intersection of the two diagonals of the feedback image, and the polar axis of the second polar coordinate system is a ray extending horizontally to the right.

[0085] Step S850: The management terminal marks the second eyeball marker in the second frame image that remains in the same position for a duration longer than a preset duration (e.g., 2 seconds) as the target eyeball marker.

[0086] Specifically, if the duration for which the manager's second eye marker remains in the same position exceeds the preset duration in the second frame image, it indicates that the manager is paying close attention to that position, thus proving that the area where that position (the position corresponding to the target eye marker in the feedback image) is located is a key area.

[0087] Step S860: The management terminal substitutes the first polar coordinate system into the second frame image to determine the polar coordinates of the target eyeball marker in the second frame image in the first polar coordinate system. .

[0088] Specifically, once the polar coordinates of the target eyeball marker in the first polar coordinate system of the second frame image are known, the position of the target eyeball marker in the second frame image can be specifically and quantitatively indicated.

[0089] Step S870: The management terminal is based on the ratio and Calculate the polar coordinates of the focal point corresponding to the second eye marker in the feedback image in the second polar coordinate system. : , In the formula, It represents a ratio.

[0090] Specifically, since the ratio can reflect the proportional relationship between the size of the manager's eye movement area and the size of the feedback image, the focal point corresponding to the second eye marker point on the feedback image can be obtained based on the above calculation formula (the polar angle remains unchanged while the polar radius changes proportionally). In other words, the focal point of the manager on the feedback image is the focal point of the second eye marker point.

[0091] Step S880: The management terminal determines the key area based on the polar coordinates of the focal point. The key area is a circular area with the focal point as the center and a third preset value (which is 20% of the length of the feedback image, for example, 200px, i.e., 200 pixels) as the radius. When there are multiple key areas and there are intersecting areas, the intersecting area is only taken once. When part of the key area exceeds the feedback image, the part that exceeds the feedback image is discarded.

[0092] In the ninth embodiment of the intelligent management method for tourism service supply chain proposed in this invention, based on the eighth embodiment, step S650 includes the following steps: Step S910: The management terminal determines the buffer area based on the polar coordinates of the focal point. The buffer area is a circular area with the focal point as the center and a fourth preset value (which is 30% of the length of the feedback image, for example, 200px, i.e., 300 pixels) as the radius. When there are multiple buffer areas and there are intersecting areas, the intersecting area is only taken once. When part of the buffer area exceeds the feedback image, the part that exceeds the feedback image is discarded. The fourth preset value is greater than the third preset value.

[0093] Step S920: The management terminal marks the area in the buffer region of the feedback image excluding the key area as the middle area, and marks the area in the feedback image other than the buffer region as the non-key area.

[0094] Step S930: The management terminal reduces the resolution of both the non-key areas and the intermediate areas to generate a simplified image, wherein the resolution of the non-key areas is lower than that of the intermediate areas.

[0095] Specifically, by setting a buffer area, which forms an intermediate area between the key area and the non-key area (less important than the key area but more important than other areas in the feedback image), the resolution of the feedback image is reduced in a stepwise manner. This preserves the readability of the feedback image to a certain extent, while preventing the boundary between the key area and the non-key area from appearing too disjointed due to the large resolution difference.

[0096] This invention also proposes a smart management system for the tourism service supply chain. The system, which applies a smart management method for the tourism service supply chain, includes a management terminal, a user terminal, and a tour guide terminal; the user terminal and the tour guide terminal are both communicatively connected to the management terminal.

[0097] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0098] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A smart management method for the tourism service supply chain, characterized in that, An application in a smart management system for the tourism service supply chain; the system includes a management terminal, a user terminal, and a tour guide terminal; both the user terminal and the tour guide terminal are communicatively connected to the management terminal; the method includes: Before the start of the tour, the management terminal will establish a correspondence between the tour guide terminal and the user terminal participating in the tour. The tour includes multiple tour components. Whenever a new tour segment is about to begin, the tour guide terminal sends the information about the upcoming tour segment to the corresponding user terminal. Each tour segment has a unique supplier. The user terminal generates feedback evaluation information corresponding to the completed tourism segments in this tour itinerary input by the user, and sends it to the management terminal. The feedback evaluation information includes evaluation type, evaluation text, and evaluation segment. The evaluation type can be any one of positive, negative, and neutral. The evaluation segment is any one of the tourism segments received from the tour guide terminal. When the number of feedback evaluation messages received by the management terminal for the same evaluation stage reaches a preset number, the supplier corresponding to the same evaluation stage is marked as the target supplier. The management terminal determines the user rating value of the target supplier based on the feedback and evaluation information of the target supplier. The higher the user rating value, the higher the tourist's satisfaction with the target supplier's service or product. The management terminal marks the target supplier whose user rating is lower than the first preset value as a supplier to be improved, generates a monitoring alarm, and sends the monitoring alarm to the supplier terminal corresponding to the supplier to be improved.

2. The intelligent management method for tourism service supply chain according to claim 1, characterized in that, Also includes: The management terminal constructs and trains a multinomial Naive Bayes model. The management terminal determines the user rating value for the target supplier based on the feedback and evaluation information corresponding to the target supplier, including: The management terminal marks the feedback evaluation information corresponding to the target supplier as target information; The management terminal performs sentiment assessments on the evaluation texts in each target information sequentially based on a trained multinomial Naive Bayes model to determine the sentiment type of the evaluation texts, wherein the sentiment type is any one of positive, negative, or neutral. The management terminal marks target information whose emotion type and evaluation type are consistent as normal information; The management terminal calculates the user rating value corresponding to the target supplier based on the target information and normal information.

3. The intelligent management method for tourism service supply chain according to claim 2, characterized in that, The formula for calculating the user rating value corresponding to the target supplier based on target information and normal information by the management terminal is as follows: ， In the formula, The user rating value corresponding to the target supplier; The number of normal information items in the target information corresponding to the target supplier; This represents the number of positive feedback ratings in the normal information.

4. The intelligent management method for tourism service supply chain according to claim 2, characterized in that, The management terminal constructs and trains a multinomial Naive Bayes model, including: The management terminal constructs a multinomial Naive Bayes model. The management terminal acquires a first text set and a second text set, wherein the first text set includes multiple first texts with a positive sentiment type, and the second text set includes multiple second texts with a negative sentiment type. The management terminal performs word segmentation on the first text set and the second text set to obtain multiple training words, and trains a multinomial Naive Bayes model based on the first text set, the second text set and each training word. The management terminal performs sentiment assessments on the evaluation texts in each target information sequentially based on a trained multinomial Naive Bayes model to determine the sentiment type of the evaluation texts, including: The management terminal sequentially performs word segmentation on the evaluation text in each target information to obtain the evaluation words corresponding to the evaluation text; The management terminal calculates the probability that the sentiment type of the evaluation text is positive and the probability that the sentiment type of the evaluation text is negative, based on the trained multinomial Naive Bayes model and each evaluation word in the evaluation text. When the probability of a text being rated as positive is greater than the probability of a text being rated as negative, the management terminal will determine the sentiment type of the text as positive. When the probability of a text being positive is less than the probability of a text being negative, the management terminal determines the sentiment of the text to be negative. When the probability that the sentiment type of the evaluation text is positive is equal to the probability that the sentiment type of the evaluation text is negative, the management terminal determines the sentiment type of the evaluation text as neutral.

5. The intelligent management method for tourism service supply chain according to claim 1, characterized in that, The feedback evaluation information also includes feedback images captured by the user through the camera module of the user terminal; the system also includes a cloud server communicatively connected to the management terminal; the management terminal also includes a display screen and a camera; the user terminal generates feedback evaluation information determined by the user and sends it to the management terminal, and then further includes: The management terminal generates a positioning page on the display screen. The positioning page has prompt text in the center and a positioning icon block in each of the four corners of the display screen. The positioning page is a rectangular page. The prompt text is used to prompt the customer to focus on each positioning icon block in turn. The management terminal acquires real-time facial video of the management personnel while the display screen shows the location page via a camera, and marks it as the first target video; The management terminal displays the feedback image on a screen and acquires real-time facial video of the management personnel while the feedback image is displayed on the screen via a camera, which is marked as the second target video. The feedback image is a rectangular image, and the size of the feedback image displayed on the screen is the same as the size of the positioning page. The management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, and the feedback image; The management terminal reduces the resolution of areas other than key areas in the feedback image to generate a simplified image; The management terminal sends the simplified image to the cloud server for storage.

6. The intelligent management method for tourism service supply chain according to claim 5, characterized in that, The management terminal determines the key areas of the feedback image based on the first target video, the second target video, and the feedback image, including: The management terminal performs image frame decomposition on the first target video to obtain multiple frame images. The obtained frame images are then mirrored and marked as the first frame image, wherein the first frame image is a rectangular image including the customer's face. The management terminal performs image recognition on each first frame image to determine the first eyeball marker point of each first frame image. The first eyeball marker point is the center point of the line segment connecting the center point of the left eyeball and the center point of the right eyeball in the first frame image. The management terminal uses the first eyeball markers closest to the four corners of the first frame image among multiple first frame images as the first target point, the second target point, the third target point, and the fourth target point, respectively. The management terminal takes the distance between the first target point and the third target point as the first distance value, and the distance between the second target point and the fourth target point as the second distance value. It takes the average of the first distance value and the second distance value and marks it as the diagonal average, where the first target point and the third target point are diagonal points, and the second target point and the fourth target point are diagonal points. The management terminal obtains the length value of any diagonal of the feedback image and marks it as the diagonal length value; it also obtains the ratio of the diagonal average value to the diagonal length value and marks it as the ratio. The management terminal determines the key area of ​​the feedback image based on the first target video, the second target video, the feedback image, and the ratio.

7. The intelligent management method for tourism service supply chain according to claim 6, characterized in that, The management terminal determines the key areas of the feedback image based on the first target video, the second target video, the feedback image, and the ratio, including: The management terminal performs image frame decomposition on the second target video to obtain multiple frame images. The obtained frame images are then mirrored and marked as the second frame image. The second frame image is a rectangular image including the customer's face, and the size of the second frame image is the same as that of the first frame image. The management terminal performs image recognition on each second frame image to determine the second eyeball marker point of each second frame image. The second eyeball marker point is the center point of the line segment connecting the center point of the left eyeball and the center point of the right eyeball in the second frame image. The management terminal constructs a first polar coordinate system based on multiple first frame images. The pole of the first polar coordinate system is the intersection of the line segment connecting the first target point and the third target point with the line segment connecting the second target point and the fourth target point. The polar axis of the first polar coordinate system is a ray extending horizontally to the right. The management terminal constructs a second polar coordinate system based on the feedback image, wherein the pole of the second polar coordinate system is the intersection of the two diagonals of the feedback image, and the polar axis of the second polar coordinate system is a ray extending horizontally to the right. The management terminal marks the second eyeball marker point in the second frame image that remains in the same position for a duration longer than a preset duration as the target eyeball marker point; The management terminal substitutes the first polar coordinate system into the second frame image to determine the polar coordinates of the target eyeball marker in the second frame image in the first polar coordinate system. ; The management terminal is based on a ratio and Calculate the polar coordinates of the focal point corresponding to the second eye marker in the feedback image in the second polar coordinate system. : ， In the formula, It is a ratio; The management terminal determines the key area based on the polar coordinates of the focal point. The key area is a circular area with the focal point as the center and a third preset value as the radius. When there are multiple key areas and they intersect, the intersecting area is taken only once. When a part of the key area exceeds the feedback image, the part that exceeds the feedback image is discarded.

8. The intelligent management method for tourism service supply chain according to claim 7, characterized in that, The management terminal reduces the resolution of areas other than key regions of the feedback image to generate a simplified image, including: The management terminal determines the buffer area based on the polar coordinates of the focal point. The buffer area is a circular area with the focal point as the center and the fourth preset value as the radius. When there are multiple buffer areas and there are intersecting areas, the intersecting area is only taken once. When part of the buffer area exceeds the feedback image, the part that exceeds the feedback image is discarded. The fourth preset value is greater than the third preset value. The management terminal marks the area outside the key area in the buffer region of the feedback image as the middle area, and marks the area outside the buffer region in the feedback image as the non-key area; The management terminal reduces the resolution of both non-critical areas and intermediate areas to generate a simplified image, wherein the resolution of non-critical areas is lower than that of intermediate areas.

9. A smart management system for the tourism service supply chain, characterized in that, The system using the intelligent management method for tourism service supply chain as described in any one of claims 1-8 includes a management terminal, a user terminal, and a tour guide terminal; the user terminal and the tour guide terminal are both communicatively connected to the management terminal.

Images