Order processing method, system and electronic equipment

Abstract

The present application provides an order processing method, including obtaining historical feature data of the current passenger; obtaining historical feature data of the plot; obtaining real-time feature data of the current passenger's order; determining a predicted order acceptance time based on the current passenger's historical feature data, the plot's historical feature data, the current passenger's real-time feature data and a preset order acceptance time prediction model; determining a display copy based on the predicted order acceptance time and preset rules; the present application can improve the prediction accuracy of the predicted order acceptance time through the order acceptance time prediction model, and by making the predicted order acceptance time transparent to the passengers, it increases the passengers' patience, reduces the passenger order cancellation rate, and further improves the passenger order success rate.

Description

Technical Field

[0001] The present invention relates to the field of transportation, and in particular to an order processing method, system and electronic equipment. Background Art

[0002] With the development of mobile Internet, the integration of traditional transportation and the Internet has flourished, and online ride-hailing has gradually become an important way for users to travel. At the same time, a way of traveling with lower costs - ride-sharing has gradually become one of the mainstream ways for users to travel.

[0003] After a passenger places an order, online ride-hailing services usually display order prediction results in the form of "expected orders within X groups" on the user's waiting order page. Order acceptance time is a very important parameter in the transaction process, which will directly affect the user's willingness to place an order and the driver-passenger matching time, and thus affect the passenger's active cancellation rate and order acceptance rate.

[0004] Rideshare orders are not dispatched by traditional online car-hailing order systems. Instead, rideshare drivers choose to take orders that match their schedules on the platform. Therefore, it is difficult to predict the exact time when a passenger will be picked up by a driver. This is a relatively complex issue, as it involves not only the supply and demand relationship of drivers and passengers who place orders at the moment, but also the number of drivers who come to the platform to look for orders.

[0005] Therefore, there is an urgent need for a method that can predict the order acceptance time to improve the order success rate, so as to solve the above-mentioned technical problems of the prior art. Summary of the invention

[0006] In order to solve the deficiencies of the prior art, the main purpose of the present invention is to provide an order processing method, system and electronic device to solve the above-mentioned technical problems of the prior art.

[0007] In order to achieve the above object, the present invention provides an order processing method in a first aspect, the method comprising:

[0008] Get the current passenger's historical characteristic data;

[0009] Obtain historical characteristic data of land parcels;

[0010] Obtain real-time feature data of current passenger orders;

[0011] Determine the predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model;

[0012] The display copy is determined based on the predicted order acceptance time and preset rules.

[0013] In some embodiments, the obtaining of historical characteristic data of a plot of land includes:

[0014] The passenger historical order information is processed according to the spatial index coding of a preset level to generate the historical characteristic data of the plot.

[0015] In some embodiments, the obtaining of the current passenger historical characteristic data includes:

[0016] Generate the passenger order acceptance portrait features and the passenger patience portrait features based on the passenger's historical order information and preset portrait generation rules;

[0017] The current passenger historical feature data is generated based on the passenger order acceptance portrait features and the passenger patience portrait features.

[0018] In some embodiments, obtaining the real-time characteristic data of the current passenger order includes:

[0019] Generate real-time order matching feature data based on real-time order information and the driver's historical order information corresponding to the driver who saw the order;

[0020] The order real-time feature data is generated according to the order real-time matching feature data and the real-time order information.

[0021] In some embodiments, before determining the predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model, the method further includes:

[0022] Filtering discrete feature data from the current passenger historical feature data, the plot historical feature data, and the current passenger order real-time feature data, and performing continuous processing on the discrete feature data to generate continuous feature data;

[0023] Filter the continuous feature data among the current passenger historical feature data, the plot historical feature data and the current passenger order real-time feature data, and standardize the continuous feature data.

[0024] In some embodiments, the method comprises:

[0025] Performing vector initialization processing on the continuous feature data after the standardization processing to obtain a vector to be processed;

[0026] Generate a first vector according to the feature display cross network and the vector to be processed;

[0027] Generate a second vector according to the feature implicit cross network and the vector to be processed;

[0028] Generate a third vector according to the shallow memory network and the vector to be processed;

[0029] The predicted order acceptance time is generated according to the first vector, the second vector, the third vector and a preset processing method.

[0030] In some embodiments, the order receiving time prediction model includes a first order receiving time prediction model and a second order receiving time prediction model, and the method further includes training the order receiving time prediction model, and the training process of the order receiving time prediction model includes:

[0031] Select the historical order information of the passenger whose departure city and arrival city are the same city as the first training sample group, wherein the first training sample group includes a first training set and a first test set;

[0032] Select the historical order information of the passenger whose departure city and arrival city are not the same city as the second training sample group, wherein the second training sample group includes a second training set and a second test set;

[0033] According to the first training set, the first order acceptance time prediction model is trained, and according to the first test set, whether the first order acceptance time prediction model meets the preset training conditions;

[0034] When the first order receiving time prediction model meets the preset training conditions, generating the first order receiving time prediction model;

[0035] According to the second training set, the second order receiving time prediction model is trained, and according to the second test set, whether the second order receiving time prediction model meets the preset training conditions;

[0036] When the second order acceptance time prediction model meets the preset training conditions, the second order acceptance time prediction model is generated.

[0037] In some embodiments, the method further comprises:

[0038] When the departure city and the destination city in the real-time characteristic data of the current passenger order are consistent, generating a first predicted order acceptance time based on the first order acceptance time prediction model;

[0039] When the departure city and the destination city in the real-time characteristic data of the current passenger order are inconsistent, generating a second predicted order acceptance time based on the second order acceptance time prediction model;

[0040] The display copy is determined according to the current time, the order issuance time, the first predicted order acceptance time, the second predicted order acceptance time and the preset rules.

[0041] In a second aspect, the present application provides an order processing system, the system comprising:

[0042] A preparation module for obtaining historical characteristic data of the current passenger;

[0043] The preparation module is also used to obtain historical characteristic data of the plot;

[0044] The preparation module is also used to obtain real-time characteristic data of the current passenger order;

[0045] A prediction module, used to determine a predicted order acceptance time according to the current passenger historical feature data, the plot historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model;

[0046] The interactive module is used to determine the display copy according to the predicted order acceptance time and preset rules.

[0047] In a third aspect, the present application provides an electronic device, the electronic device comprising:

[0048] one or more processors;

[0049] and a memory associated with the one or more processors, the memory being used to store program instructions, the program instructions when read and executed by the one or more processors performing the following operations:

[0050] Get the current passenger's historical characteristic data;

[0051] Obtain historical characteristic data of land parcels;

[0052] Obtain real-time feature data of current passenger orders;

[0053] Determine the predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model;

[0054] The display copy is determined based on the predicted order acceptance time and preset rules.

[0055] The beneficial effects achieved by this application are:

[0056] The present application provides an order processing method, including obtaining the current passenger historical feature data, plot feature data and order real-time feature data; determining the predicted order acceptance time according to the current passenger historical feature data, the plot historical feature data and the current passenger order real-time feature data and a preset order acceptance time prediction model; determining the display copy according to the predicted order acceptance time and preset rules. The present application can improve the prediction accuracy of the predicted order acceptance time through the order acceptance time prediction model, and by making the predicted order acceptance time transparent to the passengers, it increases the patience of the passengers, reduces the passenger order cancellation rate, and further improves the passenger order success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0058] Figure 1 It is a schematic diagram of an order system provided in an embodiment of the present application;

[0059] Figure 2 It is a flow chart of the order system provided in the embodiment of the present application;

[0060] Figure 3 It is a schematic diagram of an order acceptance time prediction model provided in an embodiment of the present application;

[0061] Figure 4 is a flow chart of the order processing method provided in the embodiment of the present application;

[0062] Figure 5 It is an order processing system architecture diagram provided by an embodiment of the present application;

[0063] Figure 6 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0065] Embodiment 1

[0066] refer to Figure 1 and Figure 2 The embodiment of the present application provides an order system, wherein the process of applying the order processing method disclosed in the present application to process an order includes:

[0067] S100. Obtain the historical order information of passengers on the platform offline, generate passenger order acceptance portrait features and passenger patience portrait features, that is, generate passenger historical feature data.

[0068] Specifically, in order to improve the accuracy of the passenger order acceptance portrait features, this application counts all the completed order information of the passengers on the platform (so it can be counted offline). This application counts the number of orders completed and canceled by passengers in the past week, the number of orders completed and canceled in the past two weeks, and other information, and based on the preset portrait generation rules, generates passenger order acceptance portrait features to characterize the tendency of the passenger's order to be accepted. This application generates passenger patience portrait features based on the time difference from order issuance to order cancellation in the passenger's historical order information, the event difference from order cancellation to planned departure, and other information based on the preset portrait generation rules, to characterize the passenger's patience in waiting. The current passenger historical feature data is generated in combination with the passenger order acceptance portrait features and the passenger patience portrait features.

[0069] The preset profile generation rule refers to the pre-determined generation of multiple related label information in the passenger's historical order information (such as the number of completed and cancelled orders of the passenger in the past week) and the weight coefficient of the passenger under the label, which represents the characteristic value of the passenger's feature corresponding to the label; after obtaining the weight coefficient of each label, the passenger's profile feature is obtained. Optionally, the profile feature can be in the form of a profile vector.

[0070] S200, generating historical feature data of the land parcel based on the spatial index code (H3) and the historical order information of the passengers.

[0071] Specifically, this application selects different levels based on spatial index coding to segment the service areas involved in the platform, uses a grid system to store historical order information in hexagonal plots (i.e., grids), and then generates plot feature data by counting the order waiting time for each order on the hexagon in the service area involved. Preferably, in order to improve the accuracy of the subsequent order acceptance time prediction model, this application selects spatial index codes of 6, 7, and 8 to encode and segment the service areas involved in the platform.

[0072] Among them, H3 is a hexagonal hierarchical index grid system open sourced by Uber. It is a grid-based spatial index, but unlike ordinary rectangular grid indexes, each of its grids is a regular hexagon. In grid-based spatial indexes, the more edges of the polygons used, the closer a grid is to a circle. Hexagons have the most edges and are closest to circles, so in theory they are the best choice in some scenarios. H3 abandons traditional map projections and directly covers the earth with hexagons. Although H3 cannot guarantee that each spatial unit is a hexagon, there will still be pentagons in some places, but this will not cause a big impact, because according to the characteristic that each vertex of the spherical icosahedron is in the water, this pentagon will only appear around the water and will not have a big impact on Uber's taxi and food delivery business. According to this index feature, H3 stipulates that at the 0th level of the index, each face and the upper Figure 1 In this way, there are 5.5 hexagons and 3 / 5 pentagons on each face, that is, there are 110 hexagons and 12 pentagons in total at layer 0. H3 calls these 110 hexagons base cells. H3 can reach up to 15 layers, which means that H3 has 16 levels of spatial index granularity. In the 15th layer with the finest granularity, the average size of each grid is 0.9 square meters, and the average side length is 0.509713 meters. The detailed accuracy data is shown in Table 1:

[0073] Table 1

[0074] H3 Resolution <![CDATA[Hexagon average area (km 2 )]]> Average length of hexagon (km) Number of unique indexes 0 4,250,546.8477000 1,107.712591000 122 1 607,220.9782429 418.676005500 842 2 86,745.8540347 158.244655800 5,882 3 12,392.2648621 59.810857940 41,162 4 1,770.3235517 22.606379400 288,122 5 252.9033645 8.544408276 2,016,842 6 36.1290521 3.229482772 14,117,882 7 5.1612932 1.220629759 98,825,162 8 0.7373276 0.461354684 691,776,122 9 0.1053325 0.174375668 4,842,432,842 10 0.0150475 0.065907807 33,897,029,882 11 0.0021496 0.024910561 237,279,209,162 12 0.0003071 0.009415526 1,660,954,464,122 13 0.0000439 0.003559893 11,626,681,248,842 14 0.0000063 0.001348575 81,386,768,741,882 15 0.0000009 0.000509713 569,707,381,193,162

[0075] S300. Obtain real-time characteristic data of the current passenger order.

[0076] Calculate in real time online how many drivers have seen the order initiated by the passenger, and calculate the average convenience, average starting distance, average end distance and other features of the order's historical orders corresponding to the driver who saw the order as the order's real-time matching feature data. Combine the real-time order information with the order's real-time matching feature data as the current passenger's real-time feature data.

[0077] It is worth noting that the above-mentioned step S100, step S200 and step S300 can be performed simultaneously or in sequence, and the present application does not limit the sequence.

[0078] S400: training an order acceptance time prediction model.

[0079] Since the waiting time for an order ride in the same city and across cities varies greatly, in order to more accurately predict the order receiving time and improve the targetedness of the model, the above-mentioned order receiving time prediction model is divided into a first order receiving time prediction model (i.e., a same-city order receiving time prediction model) and a second order receiving time prediction model (i.e., a cross-city order receiving time prediction model).

[0080] Preferably, the order time prediction model can be a model trained based on feature explicit cross network, feature implicit cross network, and shallow memory network. Specifically, the training process of the first order time prediction model includes:

[0081] S410, selecting historical order information of passengers whose departure city and arrival city are the same city as the first training sample, and selecting the historical order information of passengers as the first training set and the first test set according to a ratio, wherein the first training set and the first test set are generally selected at a ratio of 9:1, which is not limited in the present application;

[0082] S420: training a first order receiving time prediction model using a first training set;

[0083] S430, using the first test set, verifying whether the order acceptance time prediction accuracy of the order acceptance prediction model meets the preset conditions;

[0084] A corresponding accuracy threshold may be preset. When the prediction accuracy of the first order receiving time prediction model exceeds the accuracy threshold, it may be determined that the prediction accuracy of the first order receiving time prediction model meets the preset conditions and the first order receiving time prediction model is determined to be a trained first order receiving time prediction model.

[0085] Among them, the training process of the second order acceptance time prediction model is consistent with the above steps except that the second training sample selected is the historical order information of passengers whose departure city and arrival city are not the same city, and this application will not go into details.

[0086] S500. Determine the predicted order acceptance time according to the current passenger historical feature data, the plot historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model.

[0087] It is worth noting that before inputting the above feature data into the order acceptance time prediction model, the above feature data needs to be preprocessed.

[0088] Specifically, firstly, the discrete feature data contained in the above feature data is processed continuously. This application adopts one-hot encoding for continuous processing. Secondly, the continuous feature data is standardized, and the vector initialization processing is performed on the basis of standardization to obtain the vector to be processed, so as to facilitate the subsequent order time prediction model to predict the order time. Among them, one-hot encoding is also called unique hot encoding, which can not only solve the problem of discrete data discontinuity, but also expand the discrete data to a certain extent. It is a single-bit effective encoding, mainly using N-bit state registers to encode N states, each state has its own independent register bit, and only one bit is valid at any time. In machine learning algorithms such as regression, classification, and clustering, the calculation of the distance or similarity between features is very important. The commonly used distance or similarity calculations are similarity calculations in Euclidean space, and the calculation of cosine similarity is based on Euclidean space. The value of the feature is extended to Euclidean space, and a certain value of the discrete feature corresponds to a certain point in Euclidean space. Using one-hot encoding for discrete features will make the distance calculation between features more reasonable.

[0089] Specifically, Figure 3 As shown, the above-mentioned step of generating a predicted order acceptance time based on the order acceptance time prediction model includes:

[0090] S510, inputting the vector to be processed into a feature display cross network to generate a first vector;

[0091] Specifically, according to The operation is performed on the vector to be processed, where X0 is the vector to be processed, X1 is the first vector, W is a weight vector, and b is a bias. It is worth noting that the W value and the b value are determined during the training process of the order acceptance time prediction model.

[0092] S520, inputting the vector to be processed into a feature implicit cross network to generate a second vector;

[0093] Specifically, according to h1=ReLu(W h,0 X0+b h,0 ) operates on the vector to be processed, where X0 is the vector to be processed and h1 is the second vector.

[0094] S530, inputting the vector to be processed into the shallow memory network FM layer to generate a third vector;

[0095] Specifically, according to Operate on the vector to be processed, where X is the vector to be processed, y FM is the third vector.

[0096] S540, after splicing the first vector, the second vector and the third vector according to the weight, generate a predicted order acceptance time; specifically, according to output = W*X stack +b formula, the output is the predicted order acceptance time.

[0097] S600: Determine the display copy according to the predicted order acceptance time and preset rules.

[0098] Specifically, there are four scenarios:

[0099] 1. Just placing an order:

[0100] Assuming that the predicted time to accept an order is x minutes, and the current time is m minutes past the order issuance time, then the predicted time to accept the order is (xm) minutes. At this time, the text displayed on the passenger interface is "Start matching with car owners, expected to accept the order within (xm)."

[0101] 2. Scenario where the predicted order acceptance time is exceeded:

[0102] At this time, since the current time has exceeded the predicted order acceptance time, in order to increase the success rate of order acceptance, the passenger interface displays text such as "Try to invite the car owner to speed up order acceptance" to remind passengers to actively invite the car owner.

[0103] 3. Approaching the scheduled departure time:

[0104] At this time, the passenger interface displays text such as "As the departure time approaches, the driver is more likely to accept the order" to retain passengers.

[0105] 4. Exceeding the scheduled departure time:

[0106] At this time, the passenger interface displays text such as "You may still be accepted if you exceed the departure time, and we are continuing to match you" and similar texts, which shows passengers that the platform has been working hard to match them and improve passenger favorability.

[0107] In some implementation scenarios, the present application also proposes an experimental group and a control group to verify whether using a predictive order-pickup time model to make the order-pickup time transparent to passengers can increase the order-pickup success rate.

[0108] Specifically, the group of passengers using the order-pickup time prediction model is used as the experimental group, and the control group is the group of passengers using the default rule prompting them to accept orders (generally prompting them to accept orders within 15 minutes). Statistics and comparisons are made on a large number of data such as the order success rate and the average waiting time of passengers in the experimental group and the control group. If the experimental group data is significantly improved compared to the control group, it means that the method of using the order-pickup time prediction model to make the order-pickup time transparent to passengers can achieve the expected effect of improving the order-pickup success rate.

[0109] It should be understood that the order processing method described in this application is applied to ride-sharing order-taking platforms, online car-hailing order-taking platforms, and other platforms that provide taxi-hailing services. In particular, when applied to ride-sharing platforms, it can overcome the problem of low accuracy in predicting the order duration in existing ride-sharing order-taking platforms.

[0110] Based on the order processing method disclosed in the embodiment of the present application, the prediction accuracy of the predicted order acceptance time can be improved and the predicted order acceptance time can be made transparent to passengers, thereby increasing passengers' patience, reducing passengers' order cancellation rate, and further improving passengers' order success rate.

[0111] Embodiment 2

[0112] Corresponding to the above embodiment, the present application provides an order processing method, such as Figure 4 As shown, the method includes:

[0113] 4100. Obtaining historical characteristic data of the current passenger;

[0114] Preferably, the obtaining of the current passenger historical characteristic data includes:

[0115] 4110. Generate passenger order acceptance portrait features and passenger patience portrait features based on the passenger's historical order information and preset portrait generation rules;

[0116] 4111. Generate the current passenger historical feature data based on the passenger order acceptance portrait features and the passenger patience portrait features.

[0117] 4200. Obtain historical characteristic data of a land parcel;

[0118] Preferably, the obtaining of historical characteristic data of the land parcel includes:

[0119] 4210. Process the passenger's historical order information according to a preset level of spatial index coding to generate historical feature data of the plot of land.

[0120] 4300. Obtain real-time characteristic data of current passenger orders;

[0121] Preferably, the step of obtaining the real-time characteristic data of the current passenger order includes:

[0122] 4310. Generate real-time order matching feature data based on the real-time order information and the driver's historical order information corresponding to the driver who has seen the order;

[0123] 4311. Generate the real-time feature data of the order based on the real-time matching feature data of the order and the real-time order information.

[0124] 4400. Determine a predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data, and a preset order acceptance time prediction model;

[0125] Preferably, the predicted order acceptance time is determined based on the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data, and a preset order acceptance time prediction model:

[0126] 4410. Filter discrete feature data in the current passenger historical feature data, the plot historical feature data, and the current passenger order real-time feature data, and perform continuous processing on the discrete feature data to generate continuous feature data;

[0127] 4411. Filter the continuous feature data among the current passenger historical feature data, the plot historical feature data, and the current passenger order real-time feature data, and standardize the continuous feature data.

[0128] Preferably, the method comprises:

[0129] 4420. Perform vector initialization processing on the continuous feature data after the standardization processing to obtain a vector to be processed;

[0130] 4421. Generate a first vector according to the feature display cross network and the vector to be processed;

[0131] 4422. Generate a second vector according to the feature implicit cross network and the vector to be processed;

[0132] 4423. Generate a third vector according to the shallow memory network and the vector to be processed;

[0133] 4424. Generate the predicted order acceptance time based on the first vector, the second vector, the third vector and a preset processing method.

[0134] Preferably, the order receiving time prediction model includes a first order receiving time prediction model and a second order receiving time prediction model, and the method further includes training the order receiving time prediction model, and the training process of the order receiving time prediction model includes:

[0135] 4430. Select the historical order information of the passenger whose departure city and arrival city are the same city as a first training sample group, where the first training sample group includes a first training set and a first test set;

[0136] 4431. Select the historical order information of the passenger whose departure city and arrival city are not the same city as the second training sample group, where the second training sample group includes a second training set and a second test set;

[0137] 4432. Train the first order acceptance time prediction model according to the first training set, and verify whether the first order acceptance time prediction model meets preset training conditions according to the first test set;

[0138] 4433. When the first order receiving time prediction model meets the preset training conditions, generate the first order receiving time prediction model;

[0139] 4434. Train the second order receiving time prediction model according to the second training set, and verify whether the second order receiving time prediction model meets the preset training conditions according to the second test set.

[0140] 4435. When the second order acceptance time prediction model meets the preset training conditions, the second order acceptance time prediction model is generated.

[0141] 4500. Determine the display copy based on the predicted order acceptance time and preset rules.

[0142] Preferably, the method further comprises:

[0143] 4510. When the departure city and the destination city in the real-time characteristic data of the current passenger order are consistent, generating a first predicted order acceptance time based on the first order acceptance time prediction model;

[0144] 4511. When the departure city and the destination city in the real-time characteristic data of the current passenger order are inconsistent, generating a second predicted order acceptance time based on the second order acceptance time prediction model;

[0145] 4512. Determine the display copy according to the current time, the order issuance time, the first predicted order acceptance time, the second predicted order acceptance time and the preset rules.

[0146] Embodiment 3

[0147] refer to Figure 5 , corresponding to the above-mentioned embodiment 1 and embodiment 2, the embodiment of the present application further provides an order processing system 500, including:

[0148] A preparation module 510 is used to obtain the current passenger historical characteristic data;

[0149] The preparation module 510 is also used to obtain historical characteristic data of the land parcel;

[0150] The preparation module 510 is also used to obtain real-time characteristic data of the current passenger order;

[0151] Prediction module 520, used to determine the predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data and the current passenger order real-time feature data and a preset order acceptance time prediction model;

[0152] The interactive module 530 is used to determine the display copy according to the predicted order acceptance time and preset rules.

[0153] In some embodiments, the preparation module 510 is further used to process the passenger's historical order information according to a preset level of spatial index coding to generate the land parcel historical feature data.

[0154] In some embodiments, the preparation module 510 is also used to generate passenger order acceptance portrait features and passenger patience portrait features based on the passenger's historical order information and preset portrait generation rules; and generate the current passenger historical feature data based on the passenger order acceptance portrait features and the passenger patience portrait features.

[0155] In some embodiments, the preparation module 510 is also used to generate real-time matching feature data for the order based on the real-time order information and the driver's historical order information corresponding to the driver who saw the order; and to generate the real-time matching feature data for the order based on the real-time matching feature data for the order and the real-time order information.

[0156] In some embodiments, the preparation module 510 is also used to filter the discrete feature data in the current passenger historical feature data, the plot historical feature data and the current passenger order real-time feature data, and to perform continuous processing on the discrete feature data to generate continuous feature data; filter the continuous feature data in the current passenger historical feature data, the plot historical feature data and the current passenger order real-time feature data, and to standardize the continuous feature data.

[0157] In some embodiments, the prediction module 520 is also used to perform vector initialization processing on the continuous feature data after the standardization processing to obtain a vector to be processed; generate a first vector based on the feature explicit cross network and the vector to be processed; generate a second vector based on the feature implicit cross network and the vector to be processed; generate a third vector based on the shallow memory network and the vector to be processed; generate the predicted order acceptance time based on the first vector, the second vector and the third vector and a preset processing method.

[0158] In some embodiments, the order processing system 500 further includes a model training module 540 (not shown in the figure). The model training module 540 is used to select the historical order information of passengers whose departure city and arrival city are the same city as the first training sample group, and the first training sample group includes a first training set and a first test set; select the historical order information of passengers whose departure city and arrival city are not the same city as the second training sample group, and the second training sample group includes a second training set and a second test set; train the first order receiving time prediction model according to the first training set, and verify whether the first order receiving time prediction model meets the preset training conditions according to the first test set; generate the first order receiving time prediction model when the first order receiving time prediction model meets the preset training conditions; train the second order receiving time prediction model according to the second training set, and verify whether the second order receiving time prediction model meets the preset training conditions according to the second test set; generate the second order receiving time prediction model when the second order receiving time prediction model meets the preset training conditions.

[0159] In some embodiments, the interaction module 530 is used to generate a first predicted order acceptance time based on the first order acceptance time prediction model when the departure city and the destination city are consistent in the real-time feature data of the current passenger order; generate a second predicted order acceptance time based on the second order acceptance time prediction model when the departure city and the destination city are inconsistent in the real-time feature data of the current passenger order; determine the display copy according to the current time, the order issuance time, the first predicted order acceptance time, the second predicted order acceptance time and the preset rules.

[0160] Embodiment 4

[0161] Corresponding to all the above embodiments, an embodiment of the present application provides an electronic device, including: one or more processors; and a memory associated with the one or more processors, the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, the following operations are performed:

[0162] Get the current passenger's historical characteristic data;

[0163] Obtain historical characteristic data of land parcels;

[0164] Obtain real-time feature data of current passenger orders;

[0165] Determine the predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model;

[0166] The display copy is determined based on the predicted order acceptance time and preset rules.

[0167] in, Figure 6 The architecture of the electronic device is shown exemplarily, and may include a processor 610, a video display adapter 611, a disk drive 612, an input / output interface 613, a network interface 614, and a memory 620. The processor 610, the video display adapter 611, the disk drive 612, the input / output interface 613, the network interface 614, and the memory 620 may be communicatively connected via a bus 630.

[0168] Among them, the processor 610 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in this application.

[0169] The memory 620 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 620 can store an operating system 621 for controlling the operation of the electronic device 600, and a basic input and output system (BIOS) 622 for controlling the low-level operation of the electronic device 600. In addition, a web browser 623, a data storage management system 624, and an icon font processing system 625, etc. can also be stored. The above-mentioned icon font processing system 625 can be an application program that specifically implements the operations of the aforementioned steps in the embodiment of the present application. In short, when the technical solution provided in the present application is implemented by software or firmware, the relevant program code is stored in the memory 620 and is called and executed by the processor 610.

[0170] The input / output interface 613 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0171] The network interface 614 is used to connect to a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0172] The bus 630 comprises a pathway for transmitting information between the various components of the device (eg, the processor 610, the video display adapter 611, the disk drive 612, the input / output interface 613, the network interface 614, and the memory 620).

[0173] In addition, the electronic device 600 can also obtain information on specific collection conditions from the virtual resource object collection condition information database for use in condition judgment, etc.

[0174] It should be noted that, although the above device only shows a processor 610, a video display adapter 611, a disk drive 612, an input / output interface 613, a network interface 614, a memory 620, a bus 630, etc., in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include components necessary for implementing the solution of the present application, and does not necessarily include all the components shown in the figure.

[0175] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a cloud server, or a network device, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0176] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0177] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An order processing method, characterized in that: The method comprises: Get the current passenger's historical characteristic data; Obtain historical characteristic data of land parcels; Obtain real-time feature data of current passenger orders; Determine the predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model; Determine the display copy according to the predicted order acceptance time and preset rules; The obtaining of historical characteristic data of the plot includes: Processing the passenger's historical order information according to a preset level of spatial index coding to generate historical feature data of the land parcel; The obtaining of the current passenger historical characteristic data includes: Generate the passenger order acceptance portrait features and the passenger patience portrait features based on the passenger's historical order information and preset portrait generation rules; Generate the current passenger historical feature data according to the passenger order acceptance portrait feature and the passenger patience portrait feature; The obtaining of the real-time characteristic data of the current passenger order includes: Generate real-time order matching feature data based on real-time order information and the driver's historical order information corresponding to the driver who saw the order; Generate the order real-time feature data according to the order real-time matching feature data and the real-time order information; The order receiving time prediction model includes a first order receiving time prediction model and a second order receiving time prediction model, and the method further includes: When the departure city and the destination city in the real-time characteristic data of the current passenger order are consistent, generating a first predicted order acceptance time based on the first order acceptance time prediction model; When the departure city and the destination city in the real-time characteristic data of the current passenger order are inconsistent, generating a second predicted order acceptance time based on the second order acceptance time prediction model; The display copy is determined according to the current time, the order issuance time, the first predicted order acceptance time, the second predicted order acceptance time and the preset rules.

2. The method according to claim 1, characterized in that Before determining the predicted order acceptance time according to the current passenger historical feature data, the land parcel historical feature data, the current passenger order real-time feature data and the preset order acceptance time prediction model, the method further includes: Filtering discrete feature data from the current passenger historical feature data, the plot historical feature data, and the current passenger order real-time feature data, and performing continuous processing on the discrete feature data to generate continuous feature data; Filter the continuous feature data among the current passenger historical feature data, the plot historical feature data and the current passenger order real-time feature data, and standardize the continuous feature data.

3. The method according to claim 2, characterized in that The method comprises: Performing vector initialization processing on the continuous feature data after the standardization processing to obtain a vector to be processed; Generate a first vector according to the feature display cross network and the vector to be processed; Generate a second vector according to the feature implicit cross network and the vector to be processed; Generate a third vector according to the shallow memory network and the vector to be processed; The predicted order acceptance time is generated according to the first vector, the second vector, the third vector and a preset processing method.

4. The method according to any one of claims 1 to 3, characterized in that: The method further includes training an order receiving time prediction model, and the training process of the order receiving time prediction model includes: Select the historical order information of the passenger whose departure city and arrival city are the same city as the first training sample group, wherein the first training sample group includes a first training set and a first test set; Select the historical order information of the passenger whose departure city and arrival city are not the same city as the second training sample group, wherein the second training sample group includes a second training set and a second test set; According to the first training set, the first order acceptance time prediction model is trained, and according to the first test set, whether the first order acceptance time prediction model meets the preset training conditions; When the first order receiving time prediction model meets the preset training conditions, generating the first order receiving time prediction model; According to the second training set, the second order receiving time prediction model is trained, and according to the second test set, whether the second order receiving time prediction model meets the preset training conditions; When the second order acceptance time prediction model meets the preset training conditions, the second order acceptance time prediction model is generated.

5. An order processing system, characterized in that: The system comprises: A preparation module for obtaining historical characteristic data of the current passenger; The preparation module is also used to obtain historical characteristic data of the plot; The preparation module is also used to obtain real-time characteristic data of the current passenger order; A prediction module, used to determine a predicted order acceptance time according to the current passenger historical feature data, the plot historical feature data, the current passenger order real-time feature data and a preset order acceptance time prediction model; An interactive module, used to determine the display copy according to the predicted order acceptance time and preset rules; The preparation module is further used to process the passenger historical order information according to the spatial index coding of the preset level to generate the historical characteristic data of the land parcel; The preparation module is also used to generate passenger order acceptance portrait features and passenger patience portrait features based on passenger historical order information and preset portrait generation rules; The preparation module is also used to generate the current passenger historical feature data according to the passenger order acceptance portrait feature and the passenger patience portrait feature; The preparation module is also used to generate real-time order matching feature data based on the real-time order information and the driver's historical order information corresponding to the driver who saw the order; The preparation module is further used to generate the order real-time feature data according to the order real-time matching feature data and the real-time order information; The order-placing time prediction model includes a first order-placing time prediction model and a second order-placing time prediction model, and the interaction module is further used to generate a first predicted order-placing time based on the first order-placing time prediction model when the departure city and the destination city in the current passenger order real-time feature data are consistent; The interaction module is further configured to generate a second predicted order acceptance time based on the second order acceptance time prediction model when the departure city and the destination city in the current passenger order real-time feature data are inconsistent; The interaction module is also used to determine the display copy according to the current time, the order issuance time, the first predicted order acceptance time, the second predicted order acceptance time and the preset rules.

6. An electronic device, comprising: one or more processors; And a memory associated with the one or more processors, the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, the order processing method as described in any one of claims 1-4 is executed.

Images