Route planning method and device, equipment and computer readable storage medium

Abstract

The embodiment of the invention provides a route planning method, and the method comprises the steps: obtaining the tourism preference information of a user and the current positioning information of the user under the condition that the user selects a real-time route planning service; based on the current positioning information of the user, obtaining positioning of a plurality of sub-regions in a preset region, and determining distances between the plurality of sub-regions and the user; through a preset public opinion prediction model and a preset weather early warning model, predicting the plurality of sub-regions to obtain prediction results corresponding to the plurality of sub-regions; the preset public opinion prediction model is used for performing recommendation degree prediction on the sub-regions; the preset weather early warning model is used for performing weather prediction on the sub-regions; sorting the plurality of sub-regions based on the travel preference information of the user, the prediction results corresponding to the plurality of sub-regions and the distances between the plurality of sub-regions and the user to obtain a sub-region sorting result; generating a route track according to a sub-region sorting result; personalized route recommendation can be carried out, and the diversity of route planning is improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a route planning method and device, equipment, and computer-readable storage medium. Background Art

[0002] With the continuous development of social economy, people's demand for outdoor tourism is gradually increasing. The pursuit of tourism is not just a simple sightseeing, but they hope to enjoy the tourism deeply, meticulously and fully and appreciate the charm of nature. For large tourist attractions, there are often multiple attractions. When tourists enter the scenic area, they need to plan the travel route reasonably to maximize the use of the tour time.

[0003] The traditional route planning and scenic spot recommendation methods generally require tourists to manually plan and screen through scenic spot brochures, maps, etc., or screen according to the nearest scenic spot transportation. The planning and recommendation methods are not only inefficient, but the route planning is also relatively simple. Summary of the invention

[0004] The embodiments of the present invention are intended to provide a route planning method and apparatus, device, and computer-readable storage medium that can perform personalized route recommendations and improve the diversity of route planning.

[0005] The technical solution of the present invention is achieved in this way:

[0006] An embodiment of the present invention provides a route planning method, which includes: when a user selects a real-time route planning service, obtaining user travel preference information and user current location information; based on the user's current location information, obtaining the locations of multiple sub-areas in a preset area, and determining the distances between the multiple sub-areas and the user respectively; using a preset public opinion prediction model and a preset weather warning model, predicting the multiple sub-areas respectively, and obtaining prediction results corresponding to the multiple sub-areas respectively; the preset public opinion prediction model is used to predict the recommendation degree of the sub-areas; the preset weather warning model is used to predict the weather of the sub-areas; based on the user's travel preference information, the prediction results corresponding to the multiple sub-areas respectively, and the distances between the multiple sub-areas and the user respectively, the multiple sub-areas are sorted to obtain sub-area sorting results; and a route trajectory is generated according to the sub-area sorting results.

[0007] The embodiment of the present invention provides a route planning device, comprising: an acquisition unit, a determination unit and a generation unit, wherein:

[0008] An acquisition unit, used to acquire user travel preference information and user current location information when the user selects the real-time route planning service;

[0009] A determination unit, configured to obtain the locations of multiple sub-areas within a preset area based on the current location information of the user, and determine the distances between the multiple sub-areas and the user respectively;

[0010] The acquisition unit is further used to predict the multiple sub-regions respectively through a preset public opinion prediction model and a preset weather warning model to obtain prediction results corresponding to each of the multiple sub-regions; the preset public opinion prediction model is used to predict the recommendation degree of the sub-region; the preset weather warning model is used to predict the weather in the sub-region; based on the user's travel preference information, the prediction results corresponding to each of the multiple sub-regions, and the distances between the multiple sub-regions and the user, the multiple sub-regions are sorted to obtain the sub-region sorting results;

[0011] The generating unit is used to generate a route trajectory according to the sub-area sorting result.

[0012] An embodiment of the present invention provides a route planning device, including:

[0013] A memory for storing executable data instructions;

[0014] The processor is used to implement the above-mentioned route planning method when executing the executable instructions stored in the memory.

[0015] The embodiment of the present invention provides a route planning method and apparatus, equipment, and computer-readable storage medium, the method comprising: when a user selects a real-time route planning service, obtaining user travel preference information and user current location information; based on the user's current location information, obtaining the locations of multiple sub-areas in a preset area, and determining the distances between the multiple sub-areas and the user respectively; using a preset public opinion prediction model and a preset weather warning model, predicting the multiple sub-areas respectively, and obtaining prediction results corresponding to the multiple sub-areas; the preset public opinion prediction model is used to predict the recommendation degree of the sub-areas; the preset weather warning model is used to predict the weather for the sub-areas; based on the user's travel preference information, the prediction results corresponding to the multiple sub-areas respectively, and the distances between the multiple sub-areas and the user respectively, the multiple sub-areas are sorted to obtain sub-area sorting results; and a route trajectory is generated according to the sub-area sorting results. By adopting the above scheme, predictions can be made for multiple sub-regions according to the preset public opinion prediction model and the preset weather warning model, and prediction results corresponding to each of the multiple sub-regions can be obtained; the prediction results corresponding to each of the multiple sub-regions are combined with the user's travel preference information and the distance between the user and the multiple sub-regions to sort them, and a route trajectory is generated. Since there are deviations in the travel preferences of different users in the process of generating the route, the user's travel preference information is combined with the prediction results corresponding to each of the multiple sub-regions for route planning, which can realize personalized route recommendations and improve the diversity of route planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a physical device for route planning is provided for an embodiment of the present invention;

[0017] Figure 2 An optional process diagram of a route planning method provided in an embodiment of the present invention Figure 1 ;

[0018] Figure 3 An optional process diagram of a route planning method provided in an embodiment of the present invention Figure 2 ;

[0019] Figure 4 An optional model structure of the route planning method provided in the embodiment of the present invention Figure 1 ;

[0020] Figure 5 An optional model structure of the route planning method provided in the embodiment of the present invention Figure 2 ;

[0021] Figure 6 An optional process diagram of a route planning method provided in an embodiment of the present invention Figure 3 ;

[0022] Figure 7 An optional process diagram of a route planning method provided in an embodiment of the present invention Figure 4 ;

[0023] Figure 8 A schematic diagram of the structure of a route planning device provided by an embodiment of the present invention;

[0024] Fig. 9 A schematic diagram of the structure of a route planning device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the specific technical solution of the present invention will be further described in detail below in conjunction with the drawings in the embodiments of the present invention. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein are only for the purpose of describing the embodiments of the present invention and are not intended to limit the present invention.

[0027] In the following description, reference is made to “some embodiments”, “this embodiment”, “embodiments of the present invention” and examples, etc., which describe a subset of all possible embodiments, but it can be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0028] If similar descriptions of "first / second" appear in the application documents, the following instructions are added. In the following description, the terms "first\second\third" are only used to distinguish similar objects and do not represent a specific order of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present invention described herein can be implemented in an order other than that illustrated or described herein.

[0029] An embodiment of the present invention provides a physical device for route planning, and the physical device for route planning includes a scenic spot camera multimodal component, a cloud platform system and a terminal, as follows Figure 1 As shown,

[0030] (1) The multimodal components of the scenic spot camera include: image acquisition module, sensing module, meteorological data collection module, data processing module, information interaction and fusion module, wireless communication module, control module, equipment communication module and positioning module.

[0031] Image acquisition module: collects real-time scenic spot environment video data through the camera and uploads it to the data processing module; the image acquisition module includes an image sensor and real-time video acquisition.

[0032] Sensing module: Integrates multimodal sensors to monitor the temperature, light, humidity, and sound of scenic spots in real time, and uploads the acquired data to the data processing module.

[0033] Meteorological data collection module: used to obtain the official meteorological data of the current scenic spot and upload it to the data processing module.

[0034] Data processing module: The scenic area sets up multiple cameras, numbers the cameras and names them according to the scenic spots, decodes and adjusts the original data, filters out interference, and organizes the static data to form the real-time environmental information of the scenic spot; processes the acquired scenic spot environment video data, scenic spot temperature, light, humidity, sound data, and official meteorological data of the current scenic spot; the data processing module includes video encoding and decoding, static data processing, and data push.

[0035] Information interaction and fusion module: used to realize information interaction among the camera control module, image acquisition module, sensing module, wireless communication module, and meteorological data collection module, and to fuse the information.

[0036] Wireless communication module: used for communication between the multimodal components of the scenic spot camera and the cloud platform system, and sends the processed real-time information to the cloud platform system.

[0037] Control module: realizes the control of camera orientation, focus and other functions; wherein the control module includes an orientation angle control unit, a focus unit and a pitch angle control unit.

[0038] Device communication module: Users can remotely control the camera through the terminal device; the device communication module includes lens control components and wireless communication components.

[0039] Positioning module: used to determine the location of cameras and scenic spots.

[0040] (2) The cloud platform system includes information management module, service fee module, result output module, public opinion monitoring module, user data module, model management module and central database.

[0041] Information management module: used to manage and control devices such as cameras, as well as user management.

[0042] Service fee module: used to calculate the cost of using the service and recharge settlement; the service fee module includes a service selection module, a timer and billing module, and a settlement module.

[0043] Public opinion monitoring module: includes the public opinion capture of the current scenic spot, mainly through the location of the scenic spot, captures the data of mainstream social media, and pushes it to the central database after filtering and integration; among them, the public opinion monitoring module includes data crawling, data processing and data fusion.

[0044] User data module: includes the collection and processing of user data, and the processed data is pushed to the central database; among them, the user data module includes data collection and data processing.

[0045] Central database: used to receive real-time data, public opinion data, and user data collected from equipment, and realize the storage, processing, and integration of all data; among them, the central database includes scenic spot data storage, public opinion data storage, and central data fusion.

[0046] Model management module: training, selecting, and loading models.

[0047] Result output module: Processes and generates the results of real-time planning in the form of dynamic video, pictures, audio, and text to suit different usage scenarios, and finally transmits the results to the user terminal.

[0048] (3) User terminal: used to collect user location information in real time.

[0049] (4) User login module: used for user to select services after logging in.

[0050] In the embodiment of the present invention, the execution entity of route planning is the cloud platform system.

[0051] Figure 2 This is an optional process diagram of the route planning method provided by an embodiment of the present invention. Figure 1 , will combine Figure 2 The steps shown are explained.

[0052] S101. When a user selects a real-time route planning service, obtain user travel preference information and user current location information.

[0053] In some embodiments of the present invention, the user travel preference information includes: one of distance preference information, weather preference information and hot spot preference information.

[0054] In some embodiments of the present invention, the user's travel preference information is a travel preference questionnaire that pops up on the service selection interface after the user logs into the cloud platform system through a terminal, collecting the user's gender, age, data, travel preferences, etc. Moreover, the relevant questionnaire can be configured and modified in the user data module.

[0055] In some embodiments of the present invention, the user's current location information can be obtained through multiple cameras in the scenic spot camera multimodal component.

[0056] In some embodiments of the present invention, the cameras at the scenic spot capture pictures and real-time videos at fixed intervals. The cameras at each scenic spot are numbered, and each camera automatically takes N pictures (the number of pictures is the average of the time period) and V segments of 15s video within a set time. Each scenic spot will have a collection of pictures and videos. For example, every 10 minutes, camera 1 of scenic spot A will have 10 pictures (1 picture / 1min) and 3 segments of 15s video (1 segment / 5min, at the beginning, the 5th minute, and the 10th minute). The pictures collected by the camera at each scenic spot can be used to determine the camera where the user is currently located, and the user's current location information can be determined based on the camera number.

[0057] In some embodiments of the present invention, when entering the service selection interface, you can also select the real-time route planning service, at which time the cloud platform system performs route planning. In addition, the user can also choose not to plan the route and determine the route by obtaining information through the camera.

[0058] S102: Based on the current location information of the user, obtain the locations of multiple sub-areas in the preset area, and determine the distances between the multiple sub-areas and the user.

[0059] In some embodiments of the present invention, the cloud platform system may obtain the locations of multiple sub-areas within a preset area, and then determine the distances between the multiple sub-areas and the user respectively according to the user's current location information.

[0060] In some embodiments of the present invention, the multiple sub-area locations within the preset area may be multiple scenic spot locations within a scenic area.

[0061] In some embodiments of the present invention, the cloud platform system can obtain the location of each scenic spot based on the camera positioning of all scenic spots in the scenic area, and then determine the distance between all scenic spots and the user based on the user's current positioning information.

[0062] S103. Predictions are made for the multiple sub-regions respectively through a preset public opinion prediction model and a preset weather warning model to obtain prediction results corresponding to the multiple sub-regions; the preset public opinion prediction model is used to predict the recommendation degree of the sub-region; the preset weather warning model is used to predict the weather for the sub-region.

[0063] In some embodiments of the present invention, the preset public opinion prediction model and the preset weather warning model are both models trained in advance based on historical data and can be used directly. The preset public opinion prediction model is used to predict the recommendation degree of scenic spots in the scenic area; the preset weather warning model is used to predict the weather of scenic spots in the scenic area.

[0064] In some embodiments of the present invention, the cloud platform system obtains weather data corresponding to each of the multiple sub-regions, and public opinion data corresponding to each of the multiple sub-regions; performs public opinion predictions on the multiple sub-regions through a preset public opinion prediction model and public opinion data, and obtains public opinion prediction results corresponding to each of the multiple sub-regions; performs weather predictions on the multiple sub-regions through a preset weather warning model and weather data, and obtains weather prediction results corresponding to each of the multiple sub-regions; uses the public opinion prediction results corresponding to each of the multiple sub-regions and the weather prediction results corresponding to each of the multiple sub-regions as the prediction results corresponding to each of the multiple sub-regions.

[0065] Exemplarily, the cloud platform system obtains weather data corresponding to multiple scenic spots in the scenic area, as well as public opinion data corresponding to each of the multiple scenic spots; performs public opinion prediction on multiple scenic spots in the scenic area through a preset public opinion prediction model and public opinion data, and obtains public opinion prediction results corresponding to each of the multiple scenic spots; performs weather prediction on multiple scenic spots in the scenic area through a preset weather warning model and weather data, and obtains weather prediction results corresponding to each of the multiple scenic spots; uses the public opinion prediction results corresponding to each of the multiple scenic spots and the weather prediction results corresponding to each of the multiple scenic spots as the prediction results corresponding to each of the multiple scenic spots.

[0066] S104, sorting the multiple sub-regions based on the user's travel preference information, the prediction results corresponding to the multiple sub-regions, and the distances between the multiple sub-regions and the user to obtain a sub-region sorting result.

[0067] In some embodiments of the present invention, the user travel preference information is one of distance preference information, weather preference information and popularity preference information. If the user travel preference information is distance preference information, the cloud platform system sorts the multiple sub-regions according to the distance preference information and the distances between the multiple sub-regions and the user, and obtains the sub-region sorting result.

[0068] In some embodiments of the present invention, if the user's travel preference information is weather preference information, the cloud platform system can also sort the multiple sub-regions according to the weather preference information, the distances of the multiple sub-regions from the user, and the weather forecast results corresponding to the multiple sub-regions to obtain the sub-region sorting results.

[0069] In some embodiments of the present invention, if the user's travel preference information is popular preference information, the cloud platform system can also sort the multiple sub-regions according to the popular preference information, the distances of the multiple sub-regions from the user, and the public opinion prediction results corresponding to the multiple sub-regions to obtain the sub-region sorting results.

[0070] S105: Generate a route trajectory according to the sub-area sorting result.

[0071] In some embodiments of the present invention, the cloud platform system can generate a route track according to the sub-area sorting results and the scenic area road settings. The route track is used to guide users to visit scenic spots.

[0072] Exemplarily, the cloud platform system can obtain the travel order of the multiple attractions based on the sorting results of the multiple attractions, and generate a route trajectory based on the travel order of the multiple attractions and the road settings of the scenic area.

[0073] It can be understood that when the user selects the real-time route planning service, the user's travel preference information and the user's current location information are obtained; based on the user's current location information, the locations of multiple sub-areas in the preset area are obtained, and the distances of the multiple sub-areas from the user are determined; through the preset public opinion prediction model and the preset weather warning model, the multiple sub-areas are predicted respectively, and the prediction results corresponding to the multiple sub-areas are obtained; the preset public opinion prediction model is used to predict the recommendation degree of the sub-areas; the preset weather warning model is used to predict the weather for the sub-areas; based on the user's travel preference information, the prediction results corresponding to the multiple sub-areas and the distances of the multiple sub-areas from the user, the multiple sub-areas are sorted to obtain the sub-area sorting results; according to the sub-area sorting results, a route trajectory is generated. By adopting the above scheme, predictions can be made for multiple sub-regions according to the preset public opinion prediction model and the preset weather warning model, and prediction results corresponding to each of the multiple sub-regions can be obtained; the prediction results corresponding to each of the multiple sub-regions are combined with the user's travel preference information and the distance between the user and the multiple sub-regions to sort them, and a route trajectory is generated. Since there are deviations in the travel preferences of different users in the process of generating the route, the user's travel preference information is combined with the prediction results corresponding to each of the multiple sub-regions for route planning, which can realize personalized route recommendations and improve the diversity of route planning.

[0074] In some embodiments of the present invention, Figure 3 As shown, S103 can be implemented through S1031-S1034 as follows:

[0075] S1031. Obtain weather data corresponding to each of the plurality of sub-regions, and public opinion data corresponding to each of the plurality of sub-regions.

[0076] In some embodiments of the present invention, the weather data corresponding to each of the multiple sub-areas include: historical weather data corresponding to each of the multiple sub-areas and real-time weather data corresponding to each of the multiple sub-areas.

[0077] In some embodiments of the present invention, based on the location information of multiple sub-areas, historical data of the nearest weather station (i.e., historical weather data corresponding to each of the multiple sub-areas) is obtained, and real-time weather data (i.e., real-time weather data corresponding to each of the multiple sub-areas) is obtained through sensors.

[0078] In some embodiments of the present invention, based on the location information of multiple sub-regions, the public opinion data corresponding to each of the multiple sub-regions is obtained by crawling the data of mainstream social media through the public opinion monitoring module.

[0079] For example, social media such as XX blog, XX book, and XX audio have a wide range of users, and this type of user group is more concerned about the quality of tourism and is the main user group, so crawling data from this type of social media can better meet the needs of the user group. There are two ways to crawl data: one is to locate the address of the social media and crawl all information within a period of time with the same address as the scenic spot; the other is to filter out information with the same name as the scenic spot through keyword screening, and at the same time, take into account that different visiting times have different evaluations of the scenic spot (taking Jiuzhaigou Wucai Pool as an example, September and October are the rainy seasons, and the viewing effect is better).

[0080] It should be noted that the public opinion data are the evaluations of different users on scenic spots.

[0081] S1032. Public opinion prediction is performed on multiple sub-regions by using a preset public opinion prediction model and public opinion data to obtain public opinion prediction results corresponding to each of the multiple sub-regions.

[0082] In some embodiments of the present invention, the cloud platform system performs weather forecasts for multiple sub-regions based on the historical weather data corresponding to each of the multiple sub-regions and the real-time weather data corresponding to each of the multiple sub-regions through a preset weather warning model to obtain weather forecast results corresponding to each of the multiple sub-regions.

[0083] Exemplarily, the cloud platform system can perform weather forecasts for the multiple scenic spots based on the historical weather data corresponding to the multiple scenic spots and the real-time weather data corresponding to the multiple scenic spots through a preset weather warning model to obtain weather forecast results corresponding to the multiple scenic spots.

[0084] S1033. Perform weather forecasts on multiple sub-regions using a preset weather warning model and weather data to obtain weather forecast results corresponding to each of the multiple sub-regions.

[0085] In some embodiments of the present invention, the cloud platform system performs public opinion prediction on the multiple sub-regions based on the public opinion data corresponding to each of the multiple sub-regions through a preset public opinion prediction model to obtain public opinion prediction results corresponding to each of the multiple sub-regions.

[0086] Exemplarily, the cloud platform system can perform public opinion prediction on the multiple attractions based on the public opinion data corresponding to the multiple attractions through a preset public opinion prediction model to obtain the public opinion prediction results corresponding to the multiple attractions.

[0087] S1034. Use the public opinion prediction results corresponding to each of the multiple sub-regions and the weather prediction results corresponding to each of the multiple sub-regions as the prediction results corresponding to each of the multiple sub-regions.

[0088] In some embodiments of the present invention, the cloud platform system may use the public opinion prediction results corresponding to each of the multiple sub-regions and the weather prediction results corresponding to each of the multiple sub-regions as the prediction results corresponding to each of the multiple sub-regions.

[0089] Exemplarily, the cloud platform system may use the public opinion prediction results corresponding to the multiple scenic spots and the weather prediction results corresponding to the multiple scenic spots as the prediction results corresponding to the multiple scenic spots.

[0090] It is understandable that the cloud platform system obtains the weather data corresponding to each of the multiple sub-regions, and the public opinion data corresponding to each of the multiple sub-regions, and performs public opinion prediction on the multiple sub-regions by presetting the public opinion prediction model and public opinion data, and obtains the public opinion prediction results corresponding to each of the multiple sub-regions. By presetting the weather warning model and weather data, weather prediction is performed on the multiple sub-regions, and the weather prediction results corresponding to each of the multiple sub-regions are obtained, and the public opinion prediction results corresponding to each of the multiple sub-regions and the weather prediction results corresponding to each of the multiple sub-regions are used as the prediction results corresponding to each of the multiple sub-regions. This process fully takes into account the public opinion factors and weather factors, which is convenient for subsequent targeted route planning for users.

[0091] In some embodiments of the present invention, S104 may be implemented through S1041, or S1042, or S1043, as follows:

[0092] S1041. Sort the multiple sub-regions according to the distance preference information and the distances between the multiple sub-regions and the user, to obtain a sub-region sorting result.

[0093] In some embodiments of the present invention, the user's travel preference information is one of distance preference information, weather preference information and popularity preference information. If the user's travel preference information is distance preference information, the cloud platform system sorts the multiple sub-regions according to the distance preference information and the distances between the multiple sub-regions and the user, and obtains the sub-region sorting result.

[0094] Exemplarily, the cloud platform system sorts multiple scenic spots according to the distance preference information and the distances of the multiple scenic spots from the user, and obtains the scenic spot sorting result. As shown in Table 1, the scenic spots are Reed Sea, Shuanglong Sea, and Sparkling Sea, etc., and the distances from the user are 1km, 1.5km, and 2km respectively. The scenic spot sorting result is that Reed Sea ranks first, Shuanglong Sea ranks second, and Sparkling Sea ranks third.

[0095] Table 1

[0096] Serial number Attraction Name Distance from user 1 Reed Sea 1km 2 Shuanglong Sea 1.5km 3 sea ​​of ​​sparks 2.0km …… …… ……

[0097] S1042. Sort the multiple sub-regions according to the weather preference information, the distances between the multiple sub-regions and the user, and the weather forecast results corresponding to the multiple sub-regions to obtain a sub-region sorting result.

[0098] In some embodiments of the present invention, if the user's travel preference information is weather preference information, the cloud platform system can sort the multiple sub-regions according to the weather preference information, the distances of the multiple sub-regions from the user, and the weather forecast results corresponding to the multiple sub-regions to obtain the sub-region sorting results.

[0099] Exemplarily, the cloud platform system sorts the multiple attractions according to the weather preference information, the distances between the multiple attractions and the user, and the weather forecast results corresponding to the multiple attractions to obtain the attraction sorting result.

[0100] Table 2

[0101]

[0102] As shown in Table 2, the scenic spots are Reed Sea, Shuanglong Sea, and Sparkling Sea, etc., and the distances from the user are 1km, 1.5km, and 2km, respectively. The current weather condition of Reed Sea is heavy rain, and it will be heavy rain in the next hour; the current weather condition of Shuanglong Sea is sunny, and it will be sunny in the next hour; the current weather condition of Sparkling Sea is sunny, and it will be sunny to cloudy in the next hour; the ranking result of the scenic spots is Reed Sea third, Shuanglong Sea first, and Sparkling Sea second.

[0103] S1043. Sort the multiple sub-regions according to the popular preference information, the distances between the multiple sub-regions and the user, and the public opinion prediction results corresponding to the multiple sub-regions to obtain a sub-region sorting result.

[0104] In some embodiments of the present invention, if the user preference information is popular preference information, the cloud platform system sorts the multiple sub-regions according to the popular preference information, the distances of the multiple sub-regions from the user, and the public opinion prediction results corresponding to the multiple sub-regions to obtain the sub-region sorting results.

[0105] Exemplarily, the cloud platform system sorts multiple attractions according to the popular preference information, the distances of multiple attractions from the user, and the public opinion prediction results corresponding to the multiple attractions, and obtains the ranking results of the attractions. As shown in Table 3, the attractions are Reed Sea, Shuanglong Sea, and Sparkling Sea, etc., and the distances from the user are 1km, 1.5km, and 2km respectively. The recommendation level of Reed Sea this month is low; the recommendation level of Shuanglong Sea this month is medium; the recommendation level of Sparkling Sea this month is high; the ranking results of the attractions are Reed Sea third, Shuanglong Sea second, and Sparkling Sea first.

[0106] Table 3

[0107]

[0108] It can be understood that the cloud platform system sorts the multiple sub-regions according to the distance preference information and the distances between the multiple sub-regions and the user to obtain the sub-region sorting results; or, sorts the multiple sub-regions according to the weather preference information, the distances between the multiple sub-regions and the user and the weather forecast results corresponding to the multiple sub-regions to obtain the sub-region sorting results; or, sorts the multiple sub-regions according to the popular preference information, the distances between the multiple sub-regions and the user and the public opinion prediction results corresponding to the multiple sub-regions to obtain the sub-region sorting results, and performs personalized route planning according to different user travel preference information, thereby improving the diversity of route planning.

[0109] In some embodiments of the present invention, S105-S107 are further performed before S103, as follows:

[0110] S105: Obtain historical data of the sub-region.

[0111] In some embodiments of the present invention, the historical data includes historical sample public opinion data, and the historical data includes: first historical sample weather data and second historical sample weather data; the time corresponding to the second historical sample weather data is later than the time corresponding to the first historical sample weather data.

[0112] S106. Based on historical data, the initial public opinion prediction model is trained to determine a preset public opinion prediction model.

[0113] In some embodiments of the present invention, an initial public opinion prediction model is trained based on historical sample public opinion data to obtain a first loss value; if the first loss value is greater than or equal to a first preset loss threshold, the preset public opinion prediction model is determined; if the first loss value is less than the first preset loss threshold, training continues until the first loss value is greater than or equal to the first preset loss threshold, and the preset public opinion prediction model is determined.

[0114] For example, a public opinion prediction model for scenic spots (i.e., a preset public opinion prediction model); wherein the input of the public opinion prediction model for scenic spots is real comment data, and the output is time-scenic spot-emotion. Since there are differences in the evaluation of the same scenic spot at different times, the time dimension is introduced, and the comment data on the scenic spot from social networking sites, travel website comment areas, etc. are crawled by month. Then, the crawled data is first evaluated through crowdsourcing, and the evaluation of the scenic spot in each comment (positive, negative, neutral) is marked. The above crowd-sourced labeled data is used to build a public opinion prediction model for scenic spots based on the time scale. The specific process is as follows:

[0115] 1. Data definition: Take the real review data obtained as an example: (1) "Wuhua Lake is really beautiful in October. The colorful leaves and water are crystal clear, and it looks indescribably comfortable." (2) "On the way from Huanglong to Jiuzhaigou, I passed Ganhaizi, but the time I went there was not right. The grass and water were all dry. I don't recommend going there at the end of August." For data (1), the visited attraction is "Wuhua Lake", the visit time is October, and the emotion is marked as positive; in data (2), the visit time is the end of August, but it describes multiple attractions: Huanglong, Jiuzhaigou, and Ganhaizi. Huanglong and Jiuzhaigou are just narrative descriptions without any emotional tendency, so they are marked as neutral, while Ganhaizi should be marked as negative.

[0116] 2. Model construction

[0117] In actual scenarios, each comment has multiple times and attractions, and it is necessary to carefully align the three elements of "time-attraction-emotion". In order to solve this problem, a network model is proposed with the following structure: Figure 4 As shown: The network model includes a decoding layer, an alignment layer, and a semantic representation layer. Given a review X = {w1,...,w i ,...,w j ,...w L} , use the pre-trained model (i.e., the initial public opinion prediction model) for feature extraction, and the parameters of this layer are defined as θ f ,definition is the text feature vector obtained after the feature extraction layer, which includes time features, context features and scenic spot features. In order to identify all the time and scenic spot information in the text, an L*2 classifier is constructed, where L is used to represent each word and the two classifiers are used to distinguish time and scenic spots respectively. Figure 4 In the above, c represents the context location information other than time and scenic spots, T-0 and T-1 (not shown in the figure) represent the start and end positions of time, and S-0 and S-1 (not shown in the figure) are similar. Based on the above classifier layer, all temporal feature expressions can be obtained. Where t represents time, and i represents the i-th time in the text; h c ∈R d Similarly, for scenic spot features and context features, the correlation between time i and scenic spot j can be expressed as In the context feature h c When , the final sentiment feature is: Therefore, when there are m moments and n attractions in the text, the loss function is expressed as: The initial public opinion prediction model is trained until the loss function value (ie, the first loss value) is greater than or equal to the first preset loss threshold, and the preset public opinion prediction model is determined.

[0118] S107: Based on historical data, train the initial weather warning model to determine a preset weather warning model.

[0119] In some embodiments of the present invention, the initial weather warning model is trained based on the first historical sample weather data and the second historical sample weather data to obtain a second loss value; if the second loss value is greater than or equal to the second preset loss threshold, the preset weather warning model is determined; if the second loss value is less than the second preset loss threshold, the training continues until the second loss value is greater than or equal to the second preset loss threshold, and the preset weather warning model is determined.

[0120] For example, a weather warning model for scenic spots (i.e., a preset weather warning model); the input of the model is historical weather and real-time weather, and the output is the predicted weather at a future time. The weather change characteristics of the scenic spot are obtained through historical data, and the time and period distribution of good weather and abnormal weather at the scenic spot are obtained. Combined with the real-time data of sensors and meteorological stations, a weather warning model for scenic spots is constructed. The specific plan is divided into the following steps:

[0121] 1. Data acquisition: In order to describe the climate conditions of the scenic spot at a specific time of the day in fine granularity, the overall weather conditions of the day, the real-time sensor data at a specific time of the day, and the camera image data of the scenic spot are combined for fine-grained prediction:

[0122] (1) Numerical data: Obtain numerical data on weather conditions at the nearest meteorological point in the target scenic area over the past three years, including temperature, humidity, wind direction, wind speed, air pressure, etc. corresponding to the daily weather conditions.

[0123] (2) Fine-grained image data: crawl the image information of each scenic spot in the scenic area posted by tourists on social networking sites, travel review areas, etc. by month, and take photos of the scenic spots, while marking the specific weather conditions when taking the photos.

[0124] 2. Model construction

[0125] The overall structure is as follows Figure 5 As shown, where x 0 ,...,x n-1 represents the numerical data in the first sample, For image data, fully connected networks and convolutional neural networks are used to extract the respective features of these two different data.

[0126] The input data format is: numerical data and the corresponding crawled weather forecast data (ie, the second historical sample weather data); numerical data + image data and fine-grained weather obtained through annotation (ie, the first historical sample weather data).

[0127] The weather at a specific moment has a strong correlation with the forecast data and real-time image information, and the two tasks affect each other. On the other hand, numerical data and image data belong to two different modes. The extracted numerical features are fused with image features to make the features extracted by the common network universal, which can effectively alleviate the overfitting problem caused by a small amount of data. The constructed network is as follows: Figure 5 As shown in Figure 1, the model is mainly divided into three parts: Module 1 extracts features of original numerical data and image data. This part is divided into two sub-networks. For numerical data, a multi-layer perceptron network is used, and for image data, a convolutional neural network is used to extract. Then, a shared feature layer is used to fuse the features of the two modalities in multiple dimensions. All learnable parameters in this step are defined as θ f , can be realized by formulas (1) and (2) as follows:

[0128] G = MLP(MLP(x 0 ,...,x n-1 );CNN(x n )) (1)

[0129]

[0130] Among them, G is the vector of the fusion of numerical features and image features; MLP(x 0 ,...,x n-1 ) means extracting numerical features, and the output dimension is d_1; CNN(x n ) represents the extraction of image features, and the output dimension is d_2; ~ represents the feature fusion in splicing mode, that is, d_1 and d_2 are directly concat together, and the dimension is d_1+d_2; then MLP is used for mapping conversion, and its output dimension is d, represented by G. The x in formula 2 represents the input of the model in formula 1, that is, the fused numerical features and image features, and G_f represents Figure 5 The shared layer in is multiple MLP networks. X represents the feature vector output by the shared layer, with a dimension of d. The b in Formula 2 represents the batch size.

[0131] Module 2 is used to construct weather forecasts of different granularities. The input of this layer is the fused features obtained in module 1. In order to reduce the dimensionality difference between different modes, they are projected into a new feature space. It is necessary to connect two independent classification layers, whose parameters are defined as θ c_1 ,θ c_2 On the other hand, considering the correlation between the two tasks, certain constraints need to be imposed on the branch layer parameters: when the final predicted results are close, the difference between the parameters needs to be as small as possible, and vice versa. In order to ensure the sparsity of the parameters, a weighted The constraint is processed by regularization method, and its value is a learnable parameter. Therefore, the loss function of this module can be implemented by formula (3) and formula (4), as follows:

[0132]

[0133]

[0134] Among them, X is the output of module 1, and G_theat represents the network layer in the module, which is multiple MLPs here.

[0135] Module 3 is a classification layer that distinguishes numerical features from image features. This layer uses a multi-layer perceptron network for feature extraction. The last layer is a 2-classification layer, and the parameters of this layer are defined as θ m On the other hand, the most important function of this layer is to make the extracted features so that the classifier cannot distinguish the source of the data, indicating that the information obtained by the shared layer is completely integrated. The loss function of this layer can be implemented by formula (5) and formula (6), as follows:

[0136]

[0137]

[0138] In the specific training process, an iterative optimization method is used for training. First, the objective function is minimized to update the parameters of the label predictor, that is, according to the minimum gradient descent method parameters At the same time, in order to confuse the domain classifier, it is necessary to maximize the loss function To update the parameters At this time θ f The parameters need to be fixed.

[0139] It can be understood that the cloud platform system obtains historical data of the sub-area; based on the historical data, the initial public opinion prediction model is trained to determine the preset public opinion prediction model; based on the historical data, the initial weather warning model is trained to determine the preset weather warning model to facilitate subsequent route planning.

[0140] In an embodiment of the present invention, Figure 6 An optional process diagram of the route planning method provided by an embodiment of the present invention Figure 3 ,like Figure 6 As shown, data collection can be achieved through the following steps:

[0141] S1. Scenic spot cameras capture images and real-time videos at fixed intervals.

[0142] In some embodiments of the present invention, the cameras of each scenic spot are numbered, and each camera automatically takes N pictures (the number of pictures is the average of the time period) and V 15s videos within a set time. Each scenic spot will have a collection of pictures and videos. For example, every 10 minutes, camera 1 of scenic spot A will have 10 pictures (1 picture / 1min) and 3 15s videos (1 segment / 5min, starting, 5th minute, 10th minute).

[0143] S2. Based on the coordinates of the scenic spot, obtain the historical data of the nearest weather station, and capture real-time data through sensors to obtain meteorological data such as light, weather, noise, temperature and humidity of the current scenic spot within the set time.

[0144] S3. Based on the location of scenic spots, the public opinion monitoring module crawls the data of mainstream social media, filters it and matches the public opinion tags.

[0145] In some embodiments of the present invention, social media such as Xbo, XXshu, and Xyin have a wide range of users, and this type of user group is more concerned about the quality of tourism and sightseeing, and is the main user group of this system, so crawling data from this type of social media can better meet the needs of the user group. Data crawling will be performed in two ways, one is to locate the address carried by social media, crawl all information within a period of time with the same address as the scenic spot, and the other is to filter out information with the same name as the scenic spot by keyword screening. At the same time, considering that different visiting times have different evaluations of scenic spots (taking Jiuzhaigou Wucai Pool as an example, September and October are rainy seasons, and the viewing effect is better), so it is sorted by month during collection.

[0146] S4. The user terminal device registers user information for service purchase, and the user terminal device collects user location information in real time.

[0147] S5. The user logs in through the terminal and enters the service selection interface. A travel preference questionnaire will pop up to collect the user's gender, age, number of people, and travel preferences. The relevant questionnaire content can be configured and modified in the user data module.

[0148] It is understandable that the public opinion monitoring module in the cloud platform system can realize the collection and processing of public opinion data, and then be used to build the public opinion prediction model in the model management module: transform the crawled original comment text, introduce semantic information of the time dimension and the specific scenic spot dimension, fully tap the capabilities of the pre-trained model, and improve the accuracy of public opinion prediction.

[0149] Figure 7 An optional process diagram of a route planning method provided in an embodiment of the present invention Figure 4 , the route planning method can be implemented through the following steps, as follows:

[0150] S11: After logging in, the user selects the desired service mode: autonomously adjust the camera view, or use the real-time route planning service.

[0151] In some embodiments of the present invention, if you choose to adjust the camera yourself, you will be charged according to the user's login time. During this period, the user can choose to view the camera with different numbers at each scenic spot and can choose to remotely adjust the camera angle. If you choose to use the real-time route planning service, you do not need to pay.

[0152] S12: Select real-time route planning and fill in a questionnaire first, including collecting the user's gender, age, number of people, and travel preferences. According to the user's travel preferences in the questionnaire results, they are divided into several categories: distance preference, popular preference, and weather preference.

[0153] In some embodiments of the present invention, if the user chooses the real-time route planning service, he / she needs to fill in a questionnaire on travel preferences, including gender, age, number of people, and travel preferences. The travel preference questions include the following categories: whether to pay more attention to the distance of the scenic spot, whether to pay more attention to the popularity and popularity of the scenic spot, and whether to pay more attention to the current weather conditions of the scenic spot. According to the content of the questionnaire, the tourists are given several labels: distance preference, popularity preference, and weather preference.

[0154] S13: The system obtains the locations of all scenic spots in the scenic area according to the current location of the user, and first sorts the scenic spots according to their distance.

[0155] In some embodiments of the present invention, the system obtains the location of each scenic spot based on the camera location of all scenic spots in the scenic area, and then sorts all scenic spots from near to far according to the distance based on the user's current location to obtain a scenic spot distance list. If the user has a distance preference, the scenic spot distance sorting result is generated.

[0156] S14: The system calls the public opinion prediction model and the weather prediction model results and re-sorts them according to user tags.

[0157] S15. The sorting results are pushed to the user, supporting route navigation.

[0158] In some embodiments of the present invention, the sorted list is pushed to the user, and the user can click on the navigation to call the third-party navigation API to provide route navigation.

[0159] S16: When the user arrives at a new scenic spot, the user can obtain the latest real-time route planning service. At the same time, the user can reselect the preference settings to obtain a new route plan.

[0160] In some embodiments of the present invention, after executing S16, S14 is continued to be executed.

[0161] It is understandable that the cloud platform system uses public opinion prediction models and scenic spot weather warning models for route planning, which has the following advantages: for the public opinion model, by constructing time and scenic spot information text, it provides certain fine-grained information for public opinion prediction and improves prediction performance; for the scenic spot prediction model, it uses the information of pictures collected by scenic spot cameras combined with local weather information in a larger range to build a dual-task model, making full use of limited data for performance improvement. The method of adding camera multimodal components to upgrade the hardware of existing cameras is lower cost and requires less modification. By adding multimodal components, the existing cameras in the scenic area can be used more widely. In addition to image and video acquisition, data such as temperature, light, humidity, sound, and meteorological data collection with the location of the scenic spot as the core can also be carried out. Through the collection of multi-source data, users can be provided with more accurate route planning to meet their more detailed needs; at the same time, users can be provided with remote control camera services to make scenic route planning more personalized; in addition, the 5G communication mode is adopted to reduce the remote control delay of the camera and the image resource upload delay, and the uploaded image resources have higher clarity. The azimuth control and pitch angle control units can speed up the adjustment speed, and the focal length fine adjustment function can ensure the adjustment accuracy. For different users, manual focus or automatic focus function can be selected to further enhance the user experience. The cloud platform system has the ability to process data comprehensively, and can process, analyze, and model a variety of different data to serve more users, without restrictions on the camera and the user's location. At the same time, it is also a unified management platform that can organize and manage cameras scattered throughout the scenic area to provide services to a large number of users.

[0162] Based on the route planning method of the above embodiment, the embodiment of the present invention also provides a route planning device, such as Figure 8 As shown, Figure 8 A schematic diagram of the structure of a route planning device provided by an embodiment of the present invention is shown in FIG. Figure 8 As shown, the route planning device 8 includes: an acquisition unit 801, a determination unit 802 and a generation unit 803, wherein:

[0163] The acquisition unit 801 is used to acquire the user's travel preference information and the user's current location information when the user selects the real-time route planning service;

[0164] A determination unit 802 is used to obtain the locations of multiple sub-areas in a preset area based on the current location information of the user, and determine the distances between the multiple sub-areas and the user respectively;

[0165] The acquisition unit 801 is further used to predict the multiple sub-regions respectively through the preset public opinion prediction model and the preset weather warning model to obtain the prediction results corresponding to the multiple sub-regions; the preset public opinion prediction model is used to predict the recommendation degree of the sub-region; the preset weather warning model is used to predict the weather of the sub-region; based on the user's travel preference information, the prediction results corresponding to the multiple sub-regions, and the distances between the multiple sub-regions and the user, the multiple sub-regions are sorted to obtain the sub-region sorting results;

[0166] The generating unit 803 is used to generate a route track according to the sub-area sorting result.

[0167] In some embodiments of the present invention, the acquisition unit 801 is further used to acquire weather data corresponding to each of the multiple sub-regions, and public opinion data corresponding to each of the multiple sub-regions; perform public opinion prediction on the multiple sub-regions by using a preset public opinion prediction model and public opinion data, and obtain public opinion prediction results corresponding to each of the multiple sub-regions; perform weather prediction on the multiple sub-regions by using a preset weather warning model and weather data, and obtain weather prediction results corresponding to each of the multiple sub-regions;

[0168] The determination unit 802 is further used to use the public opinion prediction results corresponding to each of the multiple sub-regions and the weather prediction results corresponding to each of the multiple sub-regions as the prediction results corresponding to each of the multiple sub-regions.

[0169] In some embodiments of the present invention, the weather data corresponding to each of the plurality of sub-regions includes: historical weather data corresponding to each of the plurality of sub-regions and real-time weather data corresponding to each of the plurality of sub-regions;

[0170] The acquisition unit 801 is also used to perform weather forecasts on the multiple sub-regions according to the historical weather data corresponding to the multiple sub-regions and the real-time weather data corresponding to the multiple sub-regions through a preset weather warning model to obtain weather forecast results corresponding to the multiple sub-regions.

[0171] In some embodiments of the present invention, the acquisition unit 801 is also used to perform public opinion prediction on the multiple sub-regions according to the public opinion data corresponding to each of the multiple sub-regions through a preset public opinion prediction model to obtain public opinion prediction results corresponding to each of the multiple sub-regions.

[0172] In some embodiments of the present invention, the user travel preference information includes: one of distance preference information, weather preference information and hot spot preference information;

[0173] The acquisition unit 801 is further used to sort the multiple sub-regions according to the distance preference information and the distances between the multiple sub-regions and the user, and obtain the sub-region sorting result; or, sort the multiple sub-regions according to the weather preference information, the distances between the multiple sub-regions and the user, and the weather forecast results corresponding to the multiple sub-regions, and obtain the sub-region sorting result; or, sort the multiple sub-regions according to the hot preference information, the distances between the multiple sub-regions and the user, and the public opinion prediction results corresponding to the multiple sub-regions, and obtain the sub-region sorting result.

[0174] In some embodiments of the present invention, by using a preset scenic spot public opinion prediction model and a preset scenic spot weather warning model, multiple sub-areas are predicted respectively, and before the prediction results corresponding to the multiple sub-areas are obtained, the acquisition unit 801 is also used to obtain historical data of the sub-areas;

[0175] The determination unit 802 is also used to train the initial public opinion prediction model based on historical data to determine the preset public opinion prediction model; based on historical data, train the initial weather warning model to determine the preset weather warning model.

[0176] In some embodiments of the present invention, the historical data includes historical sample public opinion data;

[0177] The acquisition unit 801 is further used to train the initial public opinion prediction model according to the historical sample public opinion data to obtain a first loss value;

[0178] Determination unit 802 is also used to determine the preset public opinion prediction model if the first loss value is greater than or equal to the first preset loss threshold; if the first loss value is less than the first preset loss threshold, continue training until the first loss value is greater than or equal to the first preset loss threshold, and determine the preset public opinion prediction model.

[0179] In some embodiments of the present invention, the historical data includes: first historical sample weather data and second historical sample weather data; the time corresponding to the second historical sample weather data is later than the time corresponding to the first historical sample weather data;

[0180] The acquisition unit 801 is further used to train the initial weather warning model according to the first historical sample weather data and the second historical sample weather data to obtain a second loss value;

[0181] Determination unit 802 is also used to determine the preset weather warning model if the second loss value is greater than or equal to the second preset loss threshold; if the second loss value is less than the second preset loss threshold, continue training until the second loss value is greater than or equal to the second preset loss threshold, and determine the preset weather warning model.

[0182] Based on the route planning method of the above embodiment, the embodiment of the present invention also provides a route planning device, such as Fig. 9 As shown, Fig. 9 A structural diagram of a route planning device provided in an embodiment of the present invention, the route planning device 9 comprises: a processor 901 and a memory 902; the memory 902 stores one or more programs executable by the processor, and when one or more programs are executed, any one of the route planning methods in the previous embodiments is executed by the processor 901.

[0183] In the embodiment of the present invention, the processor 901 may be at least one of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller, and a microprocessor. It is understandable that for different devices, the electronic device used to implement the function of the processor may also be other, which is not specifically limited in the embodiment of the present invention.

[0184] Based on the route planning method of the above embodiment, an embodiment of the present invention further provides a computer-readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors. When the program is executed by the processor, the route planning method of the embodiment of the present disclosure is implemented.

[0185] Exemplarily, program instructions corresponding to a route planning method in this embodiment can be stored on a storage medium such as a CD, a hard disk, or a USB flash drive. When the program instructions corresponding to a route planning method in the storage medium are read or executed by an electronic device, the route planning method described in any of the above embodiments can be implemented.

[0186] In addition, each functional module in the embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional modules.

[0187] If the integrated unit is implemented in the form of a software function module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) or a processor to perform all or part of the steps of the method of this embodiment. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0188] It should be understood that "one embodiment" or "an embodiment" or "some embodiments" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present invention. Therefore, "in one embodiment" or "in an embodiment" or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present invention, the size of the serial number of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention. The serial numbers of the above-mentioned embodiments of the present invention are for description only and do not represent the advantages and disadvantages of the embodiments. The above description of the various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other. For the sake of brevity, this article will not go into details.

[0189] The modules described above as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules; they may be located in one place or distributed on multiple network units; some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0190] In addition, all functional modules in the embodiments of the present invention may be integrated into one processing unit, or each module may be a separate unit, or two or more modules may be integrated into one unit; the above-mentioned integrated modules may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0191] A person skilled in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM), magnetic disks or optical disks, etc., various media that can store program codes.

[0192] The methods disclosed in several method embodiments provided in the embodiments of the present invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0193] The features disclosed in several product embodiments provided in the embodiments of the present invention can be arbitrarily combined without conflict to obtain new product embodiments.

[0194] The features disclosed in several method or device embodiments provided in the embodiments of the present invention can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0195] The above is only an implementation of the embodiment of the present invention, but the protection scope of the embodiment of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the embodiment of the present invention. Therefore, the protection scope of the embodiment of the present invention should be based on the protection scope of the claims.

Claims

1. A route planning method, characterized in that: include: When the user selects the real-time route planning service, obtain the user's travel preference information and the user's current location information; Based on the current location information of the user, obtaining the locations of multiple sub-areas in a preset area, and determining the distances between the multiple sub-areas and the user respectively; By using a preset public opinion prediction model and a preset weather warning model, predictions are made for the multiple sub-regions respectively to obtain prediction results corresponding to the multiple sub-regions respectively; the preset public opinion prediction model is used to predict the recommendation degree of the sub-regions; The preset weather warning model is used to make weather forecasts for the sub-region; Based on the user travel preference information, the prediction results corresponding to the multiple sub-regions, and the distances between the multiple sub-regions and the user, the multiple sub-regions are sorted to obtain a sub-region sorting result; A route trajectory is generated according to the sub-area sorting result.

2. The method according to claim 1, characterized in that The method of performing predictions on the plurality of sub-regions respectively by using the preset public opinion prediction model and the preset weather warning model to obtain prediction results corresponding to the plurality of sub-regions respectively includes: Acquire weather data corresponding to each of the plurality of sub-regions, and public opinion data corresponding to each of the plurality of sub-regions; Using the preset public opinion prediction model and the public opinion data, public opinion prediction is performed on the multiple sub-regions to obtain public opinion prediction results corresponding to the multiple sub-regions; Performing weather forecasts on the multiple sub-regions using the preset weather warning model and the weather data to obtain weather forecast results corresponding to the multiple sub-regions; The public opinion prediction results corresponding to each of the multiple sub-regions and the weather prediction results corresponding to each of the multiple sub-regions are used as the prediction results corresponding to each of the multiple sub-regions.

3. The method according to claim 2, characterized in that The weather data corresponding to each of the plurality of sub-areas include: historical weather data corresponding to each of the plurality of sub-areas and real-time weather data corresponding to each of the plurality of sub-areas; The step of performing weather forecasting on the plurality of sub-regions by using the preset weather warning model and the weather data to obtain weather forecast results corresponding to the plurality of sub-regions respectively includes: According to the historical weather data corresponding to each of the multiple sub-regions and the real-time weather data corresponding to each of the multiple sub-regions, weather forecasts are performed on the multiple sub-regions through the preset weather warning model to obtain weather forecast results corresponding to each of the multiple sub-regions.

4. The method according to claim 2, characterized in that: The method of performing public opinion prediction on the plurality of sub-regions by using the preset public opinion prediction model and the public opinion data to obtain public opinion prediction results corresponding to each of the plurality of sub-regions includes: According to the public opinion data corresponding to each of the multiple sub-regions, public opinion predictions are performed on the multiple sub-regions using the preset public opinion prediction model to obtain public opinion prediction results corresponding to each of the multiple sub-regions.

5. The method according to any one of claims 1 to 4, characterized in that: The user travel preference information includes: one of distance preference information, weather preference information and hot spot preference information; The step of sorting the multiple sub-regions based on the user travel preference information, the prediction results corresponding to the multiple sub-regions, and the distances between the multiple sub-regions and the user to obtain the sub-region sorting result includes: sorting the multiple sub-regions according to the distance preference information and the distances between the multiple sub-regions and the user, to obtain the sub-region sorting result; or, sorting the multiple sub-areas according to the weather preference information, the distances between the multiple sub-areas and the user, and the weather forecast results corresponding to the multiple sub-areas to obtain the sub-area sorting result; or The multiple sub-regions are sorted according to the popular preference information, the distances between the multiple sub-regions and the user, and the public opinion prediction results corresponding to the multiple sub-regions, to obtain the sub-region sorting result.

6. The method according to claim 1, characterized in that Before respectively predicting the multiple sub-regions by using the preset public opinion prediction model and the preset weather warning model to obtain the prediction results corresponding to the multiple sub-regions, the method further includes: Get historical data for sub-regions; Based on the historical data, the initial public opinion prediction model is trained to determine the preset public opinion prediction model; Based on the historical data, the initial weather warning model is trained to determine the preset weather warning model.

7. The method according to claim 6, characterized in that The historical data includes historical sample public opinion data; The step of training the initial public opinion prediction model based on the historical data to determine the preset public opinion prediction model includes: According to the historical sample public opinion data, the initial public opinion prediction model is trained to obtain a first loss value; If the first loss value is greater than or equal to a first preset loss threshold, determining the preset public opinion prediction model; If the first loss value is less than the first preset loss threshold, continue training until the first loss value is greater than or equal to the first preset loss threshold, and determine the preset public opinion prediction model.

8. The method according to claim 6, characterized in that The historical data includes: first historical sample weather data and second historical sample weather data; the time corresponding to the second historical sample weather data is later than the time corresponding to the first historical sample weather data; The training of the initial weather warning model based on the historical data to determine the preset weather warning model includes: Training the initial weather warning model according to the first historical sample weather data and the second historical sample weather data to obtain a second loss value; If the second loss value is greater than or equal to a second preset loss threshold, determining the preset weather warning model; If the second loss value is less than the second preset loss threshold, continue training until the second loss value is greater than or equal to the second preset loss threshold, and determine the preset weather warning model.

9. A route planning device, characterized in that: include: an acquisition unit, a determination unit and a generation unit, wherein, The acquisition unit is used to acquire the user's travel preference information and the user's current location information when the user selects the real-time route planning service; The determining unit is configured to obtain the locations of multiple sub-areas in a preset area based on the current location information of the user, and determine the distances between the multiple sub-areas and the user respectively; The acquisition unit is further used to predict the multiple sub-regions respectively through a preset public opinion prediction model and a preset weather warning model to obtain prediction results corresponding to each of the multiple sub-regions; the preset public opinion prediction model is used to predict the recommendation degree of the sub-region; the preset weather warning model is used to predict the weather in the sub-region; based on the user's travel preference information, the prediction results corresponding to each of the multiple sub-regions, and the distances between the multiple sub-regions and the user, the multiple sub-regions are sorted to obtain a sub-region sorting result; The generating unit is used to generate a route track according to the sub-area sorting result.

10. A route planning device, characterized in that: include: A memory for storing executable data instructions; A processor, configured to implement the route planning method according to any one of claims 1 to 8 when executing the executable instructions stored in the memory.

11. A computer-readable storage medium, characterized in that: Executable instructions are stored, which are used to cause the processor to execute and implement the route planning method described in any one of claims 1 to 8.

Images