Destination recommendation method and device, computer equipment and storage medium

Abstract

The invention relates to a destination recommendation method and device, computer equipment, a storage medium and a computer program product. The method comprises the steps of obtaining a to-be-handled service type, current position information of an object, a passing tool type input by the object, destination position information corresponding to a plurality of destinations, a window number and a window processing speed; according to the current position information, the passing tool type and the position information of the multiple destinations, determining passing durations corresponding to the multiple destinations; obtaining the predicted arrival time of the object; on the basis of the predicted arrival time, the to-be-handled service type, the number of windows corresponding to the multiple destinations and the window processing speed, prediction is carried out through a waiting duration prediction model, and predicted waiting durations corresponding to the multiple destinations are obtained; and determining a recommended destination in the plurality of destinations based on the plurality of passing durations and the plurality of predicted waiting durations corresponding to the plurality of destinations. By adopting the method, the waiting time of business handling can be shortened.

Description

Technical Field

[0001] The present application relates to the field of deep learning technology, and in particular to a destination recommendation method, apparatus, computer device, storage medium and computer program product. Background Art

[0002] In order to meet the needs of the subjects for various types of business handling, each destination sets up business handling windows of various business types. There may be multiple destinations that can handle the same type of business, or only some destinations can handle specific types of business. For scenarios with multiple destinations and various business types at each destination, a destination recommendation method is urgently needed to reduce the waiting time for subjects to handle business. Summary of the invention

[0003] Based on this, it is necessary to provide a destination recommendation method, device, computer equipment, computer-readable storage medium and computer program product that can alleviate the waiting time for objects to handle business in response to the above technical problems.

[0004] In a first aspect, the present application provides a destination recommendation method, comprising:

[0005] Obtain the type of pending business, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed;

[0006] Determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the means of transportation and the location information of the multiple destinations;

[0007] Obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to multiple destinations, and the window processing speed, the waiting time prediction model is used to make predictions respectively to obtain the predicted waiting time corresponding to multiple destinations;

[0008] Based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations, a recommended destination is determined from the plurality of destinations.

[0009] In one embodiment, based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations, determining a recommended destination from the plurality of destinations includes:

[0010] For each destination, the current travel time corresponding to the current destination and the corresponding current predicted waiting time are summed to obtain the current consumption time corresponding to the current destination;

[0011] Compare the time consumption corresponding to each destination to determine the shortest time consumption;

[0012] The destination corresponding to the shortest consumed time is determined as the recommended destination.

[0013] In one embodiment, determining the travel time corresponding to each of the multiple destinations according to the current location information, the type of the travel tool, and the location information of the multiple destinations includes:

[0014] Determine multiple distance information between the current location and multiple destinations based on the current location information and multiple destination location information;

[0015] According to the type of transportation means and multiple distance information, the travel time corresponding to multiple destinations is obtained.

[0016] In one embodiment, obtaining an estimated arrival time of an object includes:

[0017] Get the current time;

[0018] Get the estimated arrival time of the object based on the current time and travel time;

[0019] Alternatively, obtain the reference arrival time of the object input and use the reference arrival time as the estimated arrival time.

[0020] In one embodiment, a method for training a waiting time prediction model includes:

[0021] Obtain a historical sample group set, the historical sample group set includes multiple historical sample groups, each historical sample group includes an arrival time sample, a service type sample, a window quantity sample, a processing speed sample, and a waiting time sample;

[0022] Inputting each arrival time sample, each service type sample, each destination window quantity sample and each processing speed sample in the historical sample group set into the initial regression model to obtain multiple intermediate prediction values;

[0023] The initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model.

[0024] In one embodiment, an initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model, including:

[0025] Based on each waiting time sample, each intermediate prediction value and the loss function, the gradient descent method is used to determine the model parameters;

[0026] Based on the initial regression model and model parameters, a waiting time prediction model is determined.

[0027] In a second aspect, the present application also provides a destination recommendation device, including:

[0028] An acquisition module, used to acquire the type of pending business, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed;

[0029] A determination module, used to determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the travel tool and the location information of the multiple destinations;

[0030] The prediction module is used to obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations, and the window processing speed, the waiting time prediction model is used to make predictions respectively to obtain the predicted waiting time corresponding to the multiple destinations;

[0031] The recommendation module is used to determine a recommended destination from a plurality of destinations based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations.

[0032] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0033] Obtain the type of pending business, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed;

[0034] Determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the means of transportation and the location information of the multiple destinations;

[0035] Obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to multiple destinations, and the window processing speed, the waiting time prediction model is used to make predictions respectively to obtain the predicted waiting time corresponding to multiple destinations;

[0036] Based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations, a recommended destination is determined from the plurality of destinations.

[0037] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0038] Obtain the type of pending business, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed;

[0039] Determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the means of transportation and the location information of the multiple destinations;

[0040] Obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to multiple destinations, and the window processing speed, the waiting time prediction model is used to make predictions respectively to obtain the predicted waiting time corresponding to multiple destinations;

[0041] Based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations, a recommended destination is determined from the plurality of destinations.

[0042] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0043] Obtain the type of pending business, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed;

[0044] Determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the means of transportation and the location information of the multiple destinations;

[0045] Obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to multiple destinations, and the window processing speed, the waiting time prediction model is used to make predictions respectively to obtain the predicted waiting time corresponding to multiple destinations;

[0046] Based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations, a recommended destination is determined from the plurality of destinations.

[0047] The above-mentioned destination recommendation method, device, computer equipment, storage medium and computer program product first determine the travel time from the current location of the object to each destination through the current location information of the object, the destination location information and the type of transportation tool, and then based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations and the window processing speed, the waiting time prediction model predicts the predicted waiting time corresponding to each of the multiple destinations. Finally, the multiple travel times and the multiple predicted waiting times corresponding to the multiple destinations are used to determine the recommended destination among the multiple destinations. Since the predicted waiting times corresponding to the multiple destinations are predicted based on the arrival time of the object and the actual conditions of the destinations, they have high accuracy. The travel times corresponding to the multiple destinations are determined by the location of the object, the transportation tool conditions and the destination location. The destination recommendation is made based on the comprehensive predicted waiting time and travel time, which is conducive to determining the destination that takes the least time and alleviating the waiting time for the object to handle business. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 A diagram of an application environment of a destination recommendation method in one embodiment;

[0050] Figure 2 A flowchart of a destination recommendation method in one embodiment;

[0051] Figure 3 is a schematic diagram of a sub-flow chart of step 208 in one embodiment;

[0052] Figure 4 is a schematic diagram of a sub-flow chart of step 204 in one embodiment;

[0053] Figure 5 A schematic diagram of the overall flow of a destination recommendation method in one embodiment;

[0054] Figure 6 is a structural block diagram of a destination recommendation device in one embodiment;

[0055] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] The destination recommendation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 obtains the type of pending business, the current location information of the object, the type of access tool input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed; according to the current location information, the type of access tool and the location information of the multiple destinations, the corresponding travel time of each of the multiple destinations is determined; the estimated arrival time of the object is obtained; based on the estimated arrival time, the type of pending business, the number of windows corresponding to each of the multiple destinations, and the window processing speed, the waiting time prediction model is used to predict respectively, and the predicted waiting time corresponding to each of the multiple destinations is obtained; based on the multiple travel times and multiple predicted waiting times corresponding to each of the multiple destinations, the recommended destination is determined from the multiple destinations. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0058] In an exemplary embodiment, Figure 2 As shown, a destination recommendation method is provided, which is applied to Figure 1 The terminal 102 in the example is used as an example to illustrate, including the following steps 202 to 208. Among them:

[0059] Step 202, obtaining the pending business type, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed.

[0060] The pending service type refers to the service type that the subject needs to handle. In the traditional method, the subject manually selects a destination that can handle the pending service type according to the service types provided by each destination.

[0061] The current location information refers to the current location information of the object, and may also refer to the location information input by the object.

[0062] The type of access means refers to the type of access means used by the subject to travel to the destination. For example, it can be a public access means type, a personal access means type, a motor vehicle type, or a non-motor vehicle type. In some embodiments, the terminal includes a human-computer interaction interface, and the subject inputs the type of pending business, current location information, and the type of access means in the human-computer interaction interface.

[0063] The destination location information refers to the location information of the destination. In some embodiments, the destination location information corresponding to each destination is different.

[0064] The number of windows refers to the number of business windows that handle business in the destination. In some embodiments, the number of windows in each destination can be the same or different.

[0065] The window processing speed refers to the average speed at which at least one service window of each destination processes services.

[0066] Step 204, determining the travel time corresponding to each of the multiple destinations based on the current location information, the type of transportation means and the location information of the multiple destinations.

[0067] The travel time refers to the time it takes for an object to reach its destination from the location indicated by its current location information. Based on the object's current location information, the type of transportation tool used by the object, and the location information of each destination, the travel time it takes for the object to reach each destination can be determined, which is helpful for determining the destination that takes the least time.

[0068] Step 206, obtaining the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations, and the window processing speed, predictions are made separately through the waiting time prediction model to obtain the predicted waiting time corresponding to the multiple destinations.

[0069] The estimated arrival time refers to the time when the object is expected to arrive at the destination. The estimated arrival time can be calculated or input by the object in the human-computer interaction interface.

[0070] The waiting time prediction model is a trained deep learning model that is used to predict the waiting time corresponding to multiple destinations.

[0071] When an object goes to a destination to handle business, it often needs to wait in line. The waiting time for different types of business is different, and the waiting time in different time periods is also different. The predicted waiting time is based on the learning of historical data by the deep learning model, and the waiting time of the current object at each destination is predicted.

[0072] The terminal inputs the estimated arrival time, the type of pending business, the number of windows corresponding to multiple destinations, and the window processing speed into the waiting time prediction model for prediction, and obtains the predicted waiting time corresponding to multiple destinations.

[0073] Step 208: Determine a recommended destination from the plurality of destinations based on the plurality of travel times and the plurality of predicted waiting times corresponding to the plurality of destinations.

[0074] Among them, before the subject goes to the destination to handle business, the time required to reach each destination can be predicted based on multiple travel times and multiple predicted waiting times corresponding to multiple destinations, thereby determining the destination that takes the least time as the recommended destination. The subject goes to the recommended destination to handle business, which helps to alleviate the waiting time for the subject to handle business.

[0075] In the above-mentioned destination recommendation method, first, the travel time from the current location of the object to each destination is determined through the current location information of the object, the destination location information and the type of transportation tool. Then, based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations and the window processing speed, the waiting time prediction model predicts the predicted waiting time corresponding to the multiple destinations. Finally, the multiple travel times and the multiple predicted waiting times corresponding to the multiple destinations are used to determine the recommended destination among the multiple destinations. Since the predicted waiting times corresponding to the multiple destinations are predicted based on the arrival time of the object and the actual conditions of the destinations, they have high accuracy. The travel times corresponding to the multiple destinations are determined by the location of the object, the transportation tool conditions and the destination location. The destination recommendation is made based on the comprehensive predicted waiting time and travel time, which is conducive to determining the destination that takes the least time and alleviating the waiting time for the object to handle business.

[0076] In an exemplary embodiment, Figure 3 As shown, step 208, based on the multiple travel times and the multiple predicted waiting times corresponding to the multiple destinations, determines a recommended destination from the multiple destinations, including steps 302 to 306. Wherein:

[0077] Step 302: For each destination, the current travel time corresponding to the current destination and the corresponding current predicted waiting time are summed to obtain the current consumption time corresponding to the current destination.

[0078] Step 304: Compare the corresponding consumption times of each destination to determine the shortest consumption time.

[0079] Step 306: determine the destination corresponding to the shortest consumed time as the recommended destination.

[0080] For each destination, the current travel time and the current predicted waiting time are summed, and the sum is used as the current consumption time corresponding to the current destination. Since the time spent on traveling to each destination is mainly spent on traveling and waiting, the current travel time and the current predicted waiting time are summed to determine a more accurate consumption time. The consumption time with the smallest value is determined among the various consumption times and is used as the shortest consumption time. The destination corresponding to the shortest consumption time is the recommended destination.

[0081] In this embodiment, the corresponding travel time and predicted waiting time of each destination are summed up respectively to obtain the corresponding consumption time. The destination corresponding to the shortest consumption time is the recommended destination. The destination with the shortest consumption time can be determined, which is helpful to alleviate the waiting time for the object to handle business.

[0082] In an exemplary embodiment, Figure 4 As shown, step 204, according to the current location information, the type of means of transportation and the location information of multiple destinations, determines the travel time corresponding to each of the multiple destinations, including steps 402 to 404. Among them:

[0083] Step 402, determining a plurality of distance information between the current position and a plurality of destinations according to the current position information and the plurality of destination position information;

[0084] Step 404, obtaining the travel time corresponding to each of the multiple destinations according to the type of transportation means and multiple distance information.

[0085] In some embodiments, the current location information and the destination location information include information such as location coordinates.

[0086] The terminal calculates the difference between the position coordinates of the current position and the position coordinates of multiple destinations, and obtains each difference as multiple distance information between the current position and the multiple destinations.

[0087] The mapping relationship between the type of passable tool and the tool speed is pre-stored in the terminal. The terminal queries the mapping relationship between the type of passable tool and the tool speed to obtain the tool speed that matches the type of passable tool.

[0088] Divide the multiple distance information by the tool speed respectively to obtain the corresponding travel time for the object to reach multiple destinations.

[0089] In this embodiment, by determining multiple distance information between the current location and multiple destinations based on the current location information and multiple destination location information, the travel time corresponding to each of the multiple destinations can be obtained based on the type of transportation tool and the multiple distance information. Based on the travel time, it is helpful to recommend a suitable destination.

[0090] In an exemplary embodiment, obtaining the estimated arrival time of the object includes: obtaining the current time; obtaining the estimated arrival time of the object based on the current time and the travel time; or obtaining a reference arrival time input by the object, and using the reference arrival time as the estimated arrival time.

[0091] The estimated arrival time of the object may be calculated or input by the object.

[0092] In the case of being obtained through calculation, the terminal obtains the current time, sums the current time and the travel time, and obtains the result as the estimated arrival time of the object.

[0093] In the case of input through the object, the object inputs the reference arrival time by operating the human-computer interaction interface, the terminal obtains the reference arrival time, and uses the reference arrival time as the estimated arrival time.

[0094] In this embodiment, the estimated arrival time of the object can be obtained through the current time and the travel time, and can also be obtained by direct input of the object. The estimated arrival time can be used to predict the waiting time at the destination, which is conducive to obtaining a recommended destination.

[0095] In an exemplary embodiment, a training method for a waiting time prediction model includes: obtaining a set of historical sample groups, the historical sample group set including multiple historical sample groups, each historical sample group including an arrival time sample, a business type sample, a window quantity sample, a processing speed sample and a waiting time sample; inputting each arrival time sample, each business type sample, each destination window quantity sample and each processing speed sample in the historical sample group set into an initial regression model to obtain multiple intermediate prediction values; training the initial regression model based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model.

[0096] The initial regression model is a deep learning model of the regression type, and the training of the initial regression model uses historical data as training samples. For example, a historical sample group set including multiple historical sample groups can be used as training samples.

[0097] The sample data in the historical sample group is the data record of the object handling the corresponding business type at each destination at a historical time. Among them, the arrival time sample is the time when the object arrives at the destination when handling business at the destination in the historical record. The business type sample is the business type handled by the object when handling business at the destination in the historical record. The window number sample is the number of business windows in the destination where the object is located when handling business at the destination in the historical record. The processing speed sample is the average processing speed of all business windows in the destination where the object is located when handling business at the destination in the historical record. The waiting time sample is the waiting time of the object starting from the destination when handling business at the destination in the historical record.

[0098] For each historical sample group in the historical sample group set, the data in the current historical sample group is input into the initial regression model to obtain the intermediate prediction value corresponding to the current historical sample group. In some embodiments, the initial regression model can be , where m represents the number of data in the historical sample group, represents the i-th data in the historical sample group, for example, represents the arrival time sample, Indicates a sample of business types, represents the number of destination window samples and represents the processing speed sample, represents the weight corresponding to the i-th data, b represents the model parameter, Represents the intermediate prediction value.

[0099] For multiple historical sample groups in the historical sample group set, multiple intermediate prediction values ​​can be obtained.

[0100] The intermediate prediction value is used to characterize the waiting time prediction value obtained by the historical sample group using the initial linear regression model. The terminal will obtain the model loss based on multiple intermediate prediction values ​​and multiple waiting time samples, so as to train the initial regression model and obtain the waiting time prediction model.

[0101] In this embodiment, the historical records of the subject handling business at various destinations are used as training samples to train the initial regression model to obtain a waiting time prediction model, which is helpful for predicting the waiting time of the subject for business handling in the future, thereby making destination recommendations.

[0102] In an exemplary embodiment, an initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model, including: determining model parameters using a gradient descent method based on each waiting time sample, each intermediate prediction value and a loss function; and determining the waiting time prediction model based on the initial regression model and the model parameters.

[0103] The terminal inputs each waiting time sample and each intermediate prediction value into the loss function, and uses the gradient descent method to obtain the model parameters. In some embodiments, the loss function can be:

[0104]

[0105] Where N represents the number of historical sample groups, represents the waiting time sample in the nth historical sample group, Represents the intermediate prediction value corresponding to the nth historical sample group, is the parameter of the loss function, represents the model loss, Used to prevent model overfitting.

[0106] The terminal brings the obtained model parameters into the initial regression model to obtain a waiting time prediction model.

[0107] In this embodiment, the model parameters are calculated through each waiting time sample, each intermediate prediction value and the loss function, which is conducive to determining the waiting time prediction model, and is conducive to predicting the waiting time of the object for business processing at future times, thereby making destination recommendations.

[0108] To explain the destination recommendation method and effect in this solution in detail, a most detailed embodiment is described below:

[0109] This is a common destination recommendation scenario in the financial field. The destination can be multiple bank branches, each of which includes multiple business windows for handling various types of business. Figure 5 The figure shows the overall process diagram of the destination recommendation method. First, a waiting time prediction model is constructed to predict the predicted waiting time of the object at each destination, then the travel time to each destination is calculated using the current location of the object, and finally the recommended destination is determined based on each predicted waiting time and each travel time.

[0110] Specifically, the object inputs the type of pending business, the current location information of the object, and the type of transportation tool used through the human-computer interaction interface. The current location information of the object can also be obtained through the global positioning system. In addition, the terminal also obtains the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed.

[0111] The terminal determines multiple distance information between the current location and multiple destinations based on the current location information and multiple destination location information, and obtains the corresponding travel time of each of the multiple destinations based on the type of travel tool and the multiple distance information.

[0112] The terminal obtains the current time, and obtains the estimated arrival time of the object based on the current time and the travel time. Alternatively, the reference arrival time input by the object is obtained, and the reference arrival time is used as the estimated arrival time. Based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations, and the window processing speed, the waiting time prediction model is used to perform predictions respectively, and the predicted waiting time corresponding to the multiple destinations is obtained.

[0113] Among them, the training method of the waiting time prediction model includes: obtaining a set of historical sample groups, the historical sample group set includes multiple historical sample groups, each historical sample group includes arrival time samples, business type samples, window quantity samples, processing speed samples and waiting time samples, and inputting each arrival time sample, each business type sample, each destination window quantity sample and each processing speed sample in the historical sample group set into the initial regression model to obtain multiple intermediate prediction values; based on each waiting time sample, each intermediate prediction value and the loss function, the gradient descent method is used to determine the model parameters, and based on the initial regression model and the model parameters, the waiting time prediction model is determined.

[0114] In some embodiments, the initial regression model can be , where m represents the number of data in the historical sample group, represents the i-th data in the historical sample group, for example, represents the arrival time sample, Indicates a sample of business types, represents the number of destination window samples and represents the processing speed sample, represents the weight corresponding to the i-th data, b represents the model parameter, Represents the intermediate prediction value.

[0115] For multiple historical sample groups in the historical sample group set, multiple intermediate prediction values ​​can be obtained. The intermediate prediction value is used to represent the waiting time prediction value obtained by the historical sample group using the initial linear regression model.

[0116] In some embodiments, the loss function may be:

[0117]

[0118] Where N represents the number of historical sample groups, represents the waiting time sample in the nth historical sample group, Represents the intermediate prediction value corresponding to the nth historical sample group, is the parameter of the loss function, represents the model loss, Used to prevent the model from overfitting. The terminal will get the model parameters Substitute a and b into the initial regression model to obtain the waiting time prediction model.

[0119] For each destination, the terminal sums the current travel time and the corresponding current predicted waiting time corresponding to the current destination to obtain the current consumption time corresponding to the current destination, compares the consumption times corresponding to each destination, determines the shortest consumption time, and determines the destination corresponding to the shortest consumption time as the recommended destination.

[0120] The above-mentioned destination recommendation method first determines the travel time from the current location of the object to each destination through the current location information of the object, the destination location information and the type of transportation tool, and then based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations and the window processing speed, the waiting time prediction model predicts the predicted waiting time corresponding to the multiple destinations. Finally, the multiple travel times and the multiple predicted waiting times corresponding to the multiple destinations are used to determine the recommended destination among the multiple destinations. Since the predicted waiting times corresponding to the multiple destinations are predicted based on the arrival time of the object and the actual conditions of the destinations, they have high accuracy. The travel times corresponding to the multiple destinations are determined by the location of the object, the transportation tool conditions and the destination location. The destination recommendation is made based on the comprehensive predicted waiting time and travel time, which is conducive to determining the destination that takes the least time and alleviating the waiting time for the object to handle business.

[0121] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0122] Based on the same inventive concept, the embodiment of the present application also provides a destination recommendation device for implementing the destination recommendation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more destination recommendation device embodiments provided below can refer to the limitations of the destination recommendation method above, and will not be repeated here.

[0123] In an exemplary embodiment, Figure 6 As shown, a destination recommendation device 600 is provided, comprising: an acquisition module 620, a determination module 640, a prediction module 660 and a recommendation module 680, wherein:

[0124] The acquisition module 620 is used to acquire the pending business type, the current location information of the object, the type of the means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed;

[0125] A determination module 640 is used to determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the travel tool and the location information of the multiple destinations;

[0126] Prediction module 660, used to obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations and the window processing speed, predictions are made through the waiting time prediction model to obtain the predicted waiting time corresponding to the multiple destinations;

[0127] The recommendation module 680 is used to determine a recommended destination from a plurality of destinations based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations.

[0128] The above-mentioned destination recommendation method first determines the travel time from the current location of the object to each destination through the current location information of the object, the destination location information and the type of transportation tool, and then based on the estimated arrival time, the type of pending business, the number of windows corresponding to the multiple destinations and the window processing speed, the waiting time prediction model predicts the predicted waiting time corresponding to the multiple destinations. Finally, the multiple travel times and the multiple predicted waiting times corresponding to the multiple destinations are used to determine the recommended destination among the multiple destinations. Since the predicted waiting times corresponding to the multiple destinations are predicted based on the arrival time of the object and the actual conditions of the destinations, they have high accuracy. The travel times corresponding to the multiple destinations are determined by the location of the object, the transportation tool conditions and the destination location. The destination recommendation is made based on the comprehensive predicted waiting time and travel time, which is conducive to determining the destination that takes the least time and alleviating the waiting time for the object to handle business.

[0129] In one embodiment, based on multiple travel times and multiple predicted waiting times corresponding to multiple destinations, a recommended destination is determined from multiple destinations, and the recommendation module 680 is also used to: for each destination, sum the current travel time and the corresponding current predicted waiting time corresponding to the current destination to obtain the current consumption time corresponding to the current destination; compare the consumption times corresponding to each destination to determine the shortest consumption time; determine the destination corresponding to the shortest consumption time as the recommended destination.

[0130] In one embodiment, the travel time corresponding to each of the multiple destinations is determined based on the current location information, the type of transportation tool and the location information of multiple destinations. The determination module 640 is also used to: determine multiple distance information between the current location and the multiple destinations based on the current location information and the location information of multiple destinations; and obtain the travel time corresponding to each of the multiple destinations based on the type of transportation tool and the multiple distance information.

[0131] In one embodiment, the estimated arrival time of the object is obtained, and the prediction module 660 is also used to: obtain the current time; obtain the estimated arrival time of the object based on the current time and the travel time; or obtain the reference arrival time input by the object, and use the reference arrival time as the estimated arrival time.

[0132] In one embodiment, in terms of training the waiting time prediction model, the prediction module 660 is also used to: obtain a set of historical sample groups, the historical sample group set includes multiple historical sample groups, each historical sample group includes an arrival time sample, a business type sample, a window quantity sample, a processing speed sample and a waiting time sample; input each arrival time sample, each business type sample, each destination window quantity sample and each processing speed sample in the historical sample group set into the initial regression model to obtain multiple intermediate prediction values; train the initial regression model based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model.

[0133] In one embodiment, the initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model. The prediction module 660 is also used to: determine the model parameters using the gradient descent method based on each waiting time sample, each intermediate prediction value and the loss function; determine the waiting time prediction model based on the initial regression model and the model parameters.

[0134] Each module in the above-mentioned destination recommendation device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0135] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 7As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a destination recommendation method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0136] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0137] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0138] Obtain the type of pending business, the current location information of the object, the type of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows and the window processing speed; determine the travel time corresponding to each of the multiple destinations based on the current location information, the type of transportation and the location information of the multiple destinations; obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to each of the multiple destinations and the window processing speed, perform predictions respectively through the waiting time prediction model to obtain the predicted waiting time corresponding to each of the multiple destinations; based on the multiple travel times and the multiple predicted waiting times corresponding to each of the multiple destinations, determine a recommended destination from the multiple destinations.

[0139] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0140] For each destination, the current travel time corresponding to the current destination and the corresponding current predicted waiting time are summed to obtain the current consumption time corresponding to the current destination; the consumption times corresponding to each destination are compared to determine the shortest consumption time; the destination corresponding to the shortest consumption time is determined as the recommended destination.

[0141] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0142] Based on the current location information and the location information of multiple destinations, multiple distance information between the current location and the multiple destinations is determined; based on the type of transportation tool and the multiple distance information, the corresponding travel time of the multiple destinations is obtained.

[0143] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0144] Get the current time; get the estimated arrival time of the object based on the current time and the travel time; or, get the reference arrival time input by the object, and use the reference arrival time as the estimated arrival time.

[0145] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0146] A historical sample group set is obtained, the historical sample group set includes multiple historical sample groups, each historical sample group includes an arrival time sample, a business type sample, a window quantity sample, a processing speed sample and a waiting time sample; each arrival time sample, each business type sample, each destination window quantity sample and each processing speed sample in the historical sample group set is input into an initial regression model to obtain multiple intermediate prediction values; the initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model.

[0147] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0148] Based on each waiting time sample, each intermediate prediction value and the loss function, the gradient descent method is used to determine the model parameters; based on the initial regression model and model parameters, the waiting time prediction model is determined.

[0149] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0150] Obtain the type of pending business, the current location information of the object, the type of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows and the window processing speed; determine the travel time corresponding to each of the multiple destinations based on the current location information, the type of transportation and the location information of the multiple destinations; obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to each of the multiple destinations and the window processing speed, perform predictions respectively through the waiting time prediction model to obtain the predicted waiting time corresponding to each of the multiple destinations; based on the multiple travel times and the multiple predicted waiting times corresponding to each of the multiple destinations, determine a recommended destination from the multiple destinations.

[0151] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0152] For each destination, the current travel time corresponding to the current destination and the corresponding current predicted waiting time are summed to obtain the current consumption time corresponding to the current destination; the consumption times corresponding to each destination are compared to determine the shortest consumption time; the destination corresponding to the shortest consumption time is determined as the recommended destination.

[0153] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0154] Based on the current location information and the location information of multiple destinations, multiple distance information between the current location and the multiple destinations is determined; based on the type of transportation tool and the multiple distance information, the corresponding travel time of the multiple destinations is obtained.

[0155] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0156] Get the current time; get the estimated arrival time of the object based on the current time and the travel time; or, get the reference arrival time input by the object, and use the reference arrival time as the estimated arrival time.

[0157] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0158] A historical sample group set is obtained, the historical sample group set includes multiple historical sample groups, each historical sample group includes an arrival time sample, a business type sample, a window quantity sample, a processing speed sample and a waiting time sample; each arrival time sample, each business type sample, each destination window quantity sample and each processing speed sample in the historical sample group set is input into an initial regression model to obtain multiple intermediate prediction values; the initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model.

[0159] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0160] Based on each waiting time sample, each intermediate prediction value and the loss function, the gradient descent method is used to determine the model parameters; based on the initial regression model and model parameters, the waiting time prediction model is determined.

[0161] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0162] Obtain the type of pending business, the current location information of the object, the type of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows and the window processing speed; determine the travel time corresponding to each of the multiple destinations based on the current location information, the type of transportation and the location information of the multiple destinations; obtain the estimated arrival time of the object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to each of the multiple destinations and the window processing speed, perform predictions respectively through the waiting time prediction model to obtain the predicted waiting time corresponding to each of the multiple destinations; based on the multiple travel times and the multiple predicted waiting times corresponding to each of the multiple destinations, determine a recommended destination from the multiple destinations.

[0163] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0164] For each destination, the current travel time corresponding to the current destination and the corresponding current predicted waiting time are summed to obtain the current consumption time corresponding to the current destination; the consumption times corresponding to each destination are compared to determine the shortest consumption time; the destination corresponding to the shortest consumption time is determined as the recommended destination.

[0165] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0166] Based on the current location information and the location information of multiple destinations, multiple distance information between the current location and the multiple destinations is determined; based on the type of transportation tool and the multiple distance information, the corresponding travel time of the multiple destinations is obtained.

[0167] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0168] Get the current time; get the estimated arrival time of the object based on the current time and the travel time; or, get the reference arrival time input by the object, and use the reference arrival time as the estimated arrival time.

[0169] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0170] A historical sample group set is obtained, the historical sample group set includes multiple historical sample groups, each historical sample group includes an arrival time sample, a business type sample, a window quantity sample, a processing speed sample and a waiting time sample; each arrival time sample, each business type sample, each destination window quantity sample and each processing speed sample in the historical sample group set is input into an initial regression model to obtain multiple intermediate prediction values; the initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model.

[0171] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0172] Based on each waiting time sample, each intermediate prediction value and the loss function, the gradient descent method is used to determine the model parameters; based on the initial regression model and model parameters, the waiting time prediction model is determined.

[0173] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0174] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0175] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0176] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A destination recommendation method, characterized in that: The method comprises: Obtain the type of pending business, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed; Determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the travel tool and the location information of the multiple destinations; Obtaining an estimated arrival time of the object; based on the estimated arrival time, the pending business type, the number of windows corresponding to the multiple destinations, and the window processing speed, respectively predicting through a waiting time prediction model to obtain a predicted waiting time corresponding to the multiple destinations; Based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations, a recommended destination is determined from the plurality of destinations.

2. The method according to claim 1, characterized in that The step of determining a recommended destination from among the multiple destinations based on the multiple travel times and the multiple predicted waiting times corresponding to the multiple destinations includes: For each destination, the current travel time corresponding to the current destination and the corresponding current predicted waiting time are summed to obtain the current consumption time corresponding to the current destination; Compare the time consumption corresponding to each destination to determine the shortest time consumption; The destination corresponding to the shortest consumed time is determined as the recommended destination.

3. The method according to claim 1, characterized in that: The determining, based on the current location information, the type of the means of transport and the location information of the multiple destinations, the travel time corresponding to each of the multiple destinations includes: Determining a plurality of distance information between the current position and a plurality of destinations according to the current position information and the plurality of destination position information; According to the type of transportation means and multiple distance information, the travel time corresponding to each of the multiple destinations is obtained.

4. The method according to claim 1, characterized in that: The obtaining of the estimated arrival time of the object includes: Get the current time; Obtaining an estimated arrival time of the object according to the current time and the travel time; Alternatively, a reference arrival time input by the subject is obtained, and the reference arrival time is used as the estimated arrival time.

5. The method according to claim 1, characterized in that The training method of the waiting time prediction model comprises: Acquire a historical sample group set, wherein the historical sample group set includes a plurality of historical sample groups, each of which includes an arrival time sample, a service type sample, a window quantity sample, a processing speed sample, and a waiting time sample; Inputting each arrival time sample, each service type sample, each destination window quantity sample and each processing speed sample in the historical sample group set into an initial regression model to obtain a plurality of intermediate prediction values; The initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model.

6. The method according to claim 5, characterized in that The initial regression model is trained based on each waiting time sample and each intermediate prediction value to obtain a waiting time prediction model, including: Based on each waiting time sample, each intermediate prediction value and the loss function, the gradient descent method is used to determine the model parameters; Based on the initial regression model and the model parameters, a waiting time prediction model is determined.

7. A destination recommendation device, characterized in that: The device comprises: An acquisition module, used to acquire the type of pending business, the current location information of the object, the type of means of transportation input by the object, the destination location information corresponding to each of the multiple destinations, the number of windows, and the window processing speed; A determination module, configured to determine the travel time corresponding to each of the multiple destinations according to the current location information, the type of the travel tool and the location information of the multiple destinations; A prediction module is used to obtain an estimated arrival time of an object; based on the estimated arrival time, the type of pending business, the number of windows corresponding to each of the multiple destinations, and the window processing speed, predictions are made through a waiting time prediction model to obtain the predicted waiting time corresponding to each of the multiple destinations; The recommendation module is used to determine a recommended destination from a plurality of destinations based on a plurality of travel times and a plurality of predicted waiting times corresponding to the plurality of destinations.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images