CN111861617A - Car pooling information processing method, computer device and computer readable storage medium - Google Patents
Car pooling information processing method, computer device and computer readable storage medium Download PDFInfo
- Publication number
- CN111861617A CN111861617A CN201910959634.2A CN201910959634A CN111861617A CN 111861617 A CN111861617 A CN 111861617A CN 201910959634 A CN201910959634 A CN 201910959634A CN 111861617 A CN111861617 A CN 111861617A
- Authority
- CN
- China
- Prior art keywords
- information
- carpooling
- historical
- degree
- order information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000011176 pooling Methods 0.000 title claims abstract description 80
- 230000010365 information processing Effects 0.000 title claims abstract description 54
- 238000003672 processing method Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000013135 deep learning Methods 0.000 claims description 6
- 230000003442 weekly effect Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 description 33
- 238000010586 diagram Methods 0.000 description 12
- 238000004364 calculation method Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000007477 logistic regression Methods 0.000 description 3
- 230000004069 differentiation Effects 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000000835 fiber Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0645—Rental transactions; Leasing transactions
-
- G06Q50/40—
Abstract
The embodiment of the disclosure provides a car pooling information processing method, computer equipment and a computer readable storage medium, wherein the car pooling information processing method comprises the following steps: counting the incidence relation among historical carpool order information, historical user portrait information and historical carpool experience information, wherein the historical carpool experience information comprises forward data and backward data; receiving carpooling order information; and predicting the carpooling experience information corresponding to the carpooling order information according to the carpooling order information, the zeroed user portrait information and the association relation. The car pooling information processing method provided by the embodiment of the disclosure counts the incidence relation among the historical car pooling order information, the historical user portrait information and the historical car pooling experience information in the historical data, and improves the accuracy of prediction. When the incidence relation is used, the user portrait information is set to be zero only by adopting the car pooling order information, and the user portrait information is abandoned, so that the killing can be avoided, and the killing can be avoided while the prediction accuracy is improved.
Description
Technical Field
The embodiment of the disclosure relates to the technical field of Internet taxi booking, in particular to a taxi pooling information processing method, computer equipment and a computer readable storage medium.
Background
Under the car sharing scene, the platform can filter the order that the piece together is not good through formulating some rules for improving car sharing efficiency. However, because the rules are hard and there is a false damage, some potential good spliced orders can be filtered out, so in order to improve the order splicing rate, some potential good spliced orders need to be recalled again. The existing recall method is to establish an experience recall model, predict whether a carpool order worth the second recall is in the way when the model is applied on line, and recall the carpool order into a sharing logic if the order is in the way.
However, whether the road is followed or not is a subjective index, and is related to the mood, the urgency, the sensitivity of traffic congestion around the road and the like of a user, the existing model only uses the related characteristics of order dimensionality and has deviation, so that the accuracy of predicting whether the road is followed is about seventy percent, and the recall accuracy is reduced.
Disclosure of Invention
The disclosed embodiments are directed to solving at least one of the technical problems of the related art or the related art.
To this end, a first aspect of the embodiments of the present disclosure is to provide a car pooling information processing method.
A second aspect of the embodiments of the present disclosure is to provide a computer device.
A third aspect of the embodiments of the present disclosure is to provide a computer-readable storage medium.
In view of this, according to a first aspect of the embodiments of the present disclosure, there is provided a car pooling information processing method, including: counting the incidence relation among historical carpool order information, historical user portrait information and historical carpool experience information, wherein the historical carpool experience information comprises forward data and backward data; receiving carpooling order information; and predicting the carpooling experience information corresponding to the carpooling order information according to the carpooling order information, the zeroed user portrait information and the association relation.
According to the car sharing information processing method provided by the embodiment of the disclosure, the incidence relation among the historical car sharing order information, the historical user portrait information and the historical car sharing experience information in the historical data is counted, and on the basis of predicting whether the car sharing is right-way or not based on the objective car sharing order information, the user portrait information is introduced to eliminate the bias of the car sharing experience information on different individuals (people) possibly brought by subjective characteristics, so that the weights of different data in the historical car sharing order information are more reasonable when the incidence relation is counted, and the prediction accuracy is improved. When the incidence relation is used, the user portrait information is set to be zero only by adopting the car pooling order information, and the user portrait information is abandoned, so that the killing can be avoided, and the killing can be avoided while the prediction accuracy is improved. In addition, the user portrait information is set to zero when the incidence relation is used, the data processing amount during calculation can be reduced, and the hardware operation load is reduced, so that the hardware requirement for predicting the carpooling experience by using the incidence relation can be reduced, the calculation efficiency can be improved, and the calculation time is shortened.
In addition, according to the car pooling information processing method in the above technical solution provided by the embodiment of the present disclosure, the following additional technical features may also be provided:
in the above technical solution, preferably, the step of predicting the carpooling experience information corresponding to the carpooling order information according to the carpooling order information, the zeroed user portrait information and the association relationship includes: determining the degree of following corresponding to the carpool order information according to the carpool order information, the zeroed user portrait information and the association relation; and predicting the carpooling experience information corresponding to the carpooling order information according to the road degree.
In the technical scheme, how to use the incidence relation obtained by statistics to predict the carpooling experience information is specifically limited. The method comprises the steps of firstly substituting car pooling order information and zeroed user portrait information into an incidence relation to obtain the following degree of the car pooling order information so as to reflect the following degree of the order, then predicting car pooling experience information according to the following degree, quantifying the following degree and improving the prediction precision.
In any of the above technical solutions, preferably, the step of predicting the carpooling experience information corresponding to the carpooling order information according to the degree of the way includes: determining that the degree of following is greater than or equal to a threshold value of the degree of following, and predicting the carpooling experience information corresponding to the carpooling order information as the data of following; and determining that the degree of following is smaller than the threshold value of the degree of following, and predicting the carpooling experience information corresponding to the carpooling order information as the out-of-way data.
In the technical scheme, how to predict the carpooling experience information according to the degree of the way is specifically limited. The forward degree threshold value is set to represent the forward degree during forward road, when the forward degree is larger than or equal to the forward degree threshold value, the carpooling experience information can be considered as forward data, otherwise, when the forward degree is smaller than the forward degree threshold value, the carpooling experience information is considered as out-of-road data, on one hand, clear judgment standards can be provided, on the other hand, the judgment standards during prediction can be changed by adjusting the forward degree threshold value, the prediction flexibility is improved, and the method is suitable for various scenes.
In any one of the above technical solutions, preferably, the car pooling information processing method further includes: and determining a forward degree threshold according to the accuracy of the historical predicted forward degree determined by the application incidence relation.
In the technical scheme, the car sharing information processing method is further limited to determine the degree of following road threshold. After the incidence relation obtained by statistics is applied to the carpooling order information, the corresponding road following degree can be predicted, and when the order corresponding to the carpooling order information is completed, the actual road following condition of the order can be obtained, so that whether the prediction result is accurate or not can be known. After a certain amount of application history data obtained by applying the incidence relation is accumulated, the accuracy of historical road degree prediction can be counted, and a reasonable road degree threshold value is determined according to the accuracy, so that the prediction accuracy is improved.
In any of the above technical solutions, preferably, the step of counting an association relationship between the historical carpooling order information, the historical user portrait information, and the historical carpooling experience information includes: and carrying out statistics on the association relationship among the historical carpooling order information, the historical user portrait information, the historical carpooling experience information and the road following degree by using a deep learning algorithm.
In the technical scheme, a scheme of statistical association is specifically defined. The association relationship is counted by using a deep learning algorithm, that is, a prediction model (which can be a general model) is firstly created, and the model is trained by using historical data, so that the prediction model capable of reflecting the association relationship can be obtained. Specifically, when the model is trained, the historical car pooling order information and the historical user portrait information are used as training parameters, the historical car pooling experience information is used as a mark, the degree of the way is used as a target parameter, and training is performed, so that the prediction model can naturally embody the reasonable weight relationship between the car pooling order information and the user portrait information, and the model training process is facilitated to be simplified. When the model is applied, the carpool order information is input into the model, the user image information in the model is set to zero, namely the weight is removed, and the prejudice possibly brought by the main factor is eliminated by the output degree of the way.
In any of the above technical solutions, preferably, the step of counting an association relationship between the historical carpooling order information, the historical user portrait information, and the historical carpooling experience information includes: establishing a first incidence relation among historical carpool order information, historical carpool experience information and a first road degree; creating a second association relation among historical user portrait information, historical carpooling experience information and a second road degree; and performing joint statistics on the first association relation and the second association relation by using the same historical car pooling order information, historical user portrait information and historical car pooling experience information.
In the technical scheme, another scheme of statistical association relationship is specifically defined. Respectively establishing an association relation with the degree of following the road aiming at the historical carpool order information and the historical user portrait information, namely a first association relation and a second association relation, performing combined statistics on the first association relation and the second association relation by using the same historical data during statistics to associate the first association relation and the second association relation, wherein the statistical results can respectively reflect the influence of two factors on the degree of following the road, and the differentiation and association are realized.
In any of the above technical solutions, preferably, the step of determining the degree of following route corresponding to the car pool order information according to the car pool order information, the zeroed user portrait information, and the association relationship includes: and determining a first road degree according to the car pooling order information and the first association relation, and taking the first road degree as the road degree corresponding to the car pooling order information.
In the technical scheme, how to apply the incidence relation to predict the degree of forward road is specifically limited when the second scheme is adopted to count the incidence relation. According to the scheme of the embodiment of the disclosure, the user portrait information is abandoned when the association relationship is applied, so that the car pool order information can be directly substituted into the first association relationship, and the output first road degree reflects the influence of the user portrait information and is a result of prediction only by using the car pool order information but not the user portrait information, so that the prediction accuracy is improved and the killing is avoided. In addition, because only the first association relation is used, only the first association relation can be stored and operated on hardware, and the storage and calculation load is reduced.
In any of the above technical solutions, preferably, the performing, by a joint statistic operation, an operation on the first association relationship and the second association relationship includes: and performing joint statistics on the first association relation and the second association relation in an addition or multiplication mode.
In the technical scheme, a mode of performing joint statistics on the first incidence relation and the second incidence relation is specifically defined. The method can adopt an addition mode, namely weights are respectively given to the first road degree and the second road degree during statistics, so that the weighted value of the first road degree and the second road degree is 1; it is also possible to use a multiplication method, i.e. to pre-multiply the two output results first and then normalize them during statistics. Both of these ways can achieve joint statistics.
In any of the above technical solutions, preferably, the carpool order information includes at least one of: the method comprises the steps of ordering original time length, ordering original length, ordering estimated time length, ordering estimated length, estimated detour distance, estimated detour time, estimated driving receiving time, estimated departure time and travel scene; and/or the historical carpool order information includes at least one of: the order original time length, the order original length, the order car sharing time length, the order car sharing length, the detour distance, the detour time, the driving receiving time, the departure time and the travel scene.
In the technical scheme, data contained in the order carpooling information and the historical order carpooling information are specifically limited, and the data can be regarded as characteristics of order dimensions. By introducing the characteristics, the situation of the current carpool order can be fully known so as to predict carpool experience information. It should be noted that, the travel scene is related to the destination, time, and other factors, and may be mined by a special algorithm, which is not described in detail herein. The travel requirements under different scenes are different, and the prediction result can be adjusted accordingly, so that the prediction flexibility and the adaptability are improved.
In any of the above technical solutions, preferably, the user portrait information includes at least one of: gender, age, occupation, monthly outstanding passenger inventory, weekly driver inventory, driver rating, historical passenger feedback inventory, driver cancellation rate, driver response rate, and passenger cancellation rate.
In the technical scheme, data contained in user portrait information are specifically limited, and the data can be regarded as portrait characteristics, wherein the user refers to a passenger and a driver. By introducing the characteristics, the subjective emotion of the user can be fully reflected, and the accuracy of prediction is improved. In consideration of the fact that the frequency of orders sent by passengers is far lower than the frequency of orders received by a driver, the monthly finished-order quantity of the passengers and the weekly finished-order quantity of the drivers are respectively used in the finished-order quantity so as to ensure the balance of the collected data.
In any one of the above technical solutions, preferably, the step of receiving the carpool order information includes: receiving at least two pieces of carpool order information of a carpool to be carpooled; the car pooling information processing method further comprises the following steps: determining that the carpooling experience information corresponding to all the carpooling order information to be carpooled is the forward data, and associating all the carpooling order information to the same driver account.
In the technical scheme, a scheme of completing car sharing by using a prediction result is specifically limited. The background can receive the carpooling requests sent by different passenger users, which can be newly sent by the passengers or filtered carpooling requests, the requests include basic planned starting time and starting and ending points, when the carpooling operation is executed or whether the filtered carpooling requests need to be recalled, whether a plurality of carpooling requests to be carpooled are in-route or not needs to be calculated, at the moment, a driving route can be planned based on the carpooling requests, carpooling order information corresponding to the carpooling requests is generated, carpooling experience information is respectively predicted for each piece of carpooling order information, when the predicted carpooling experience information is all in-route data, the carpooling requests are determined or recalled, carpooling is carried out, and the carpooling order information of the determined carpooling is related to the same account number, and the carpooling operation is completed. The scheme can ensure the carpooling experience of the user.
In any of the above technical solutions, preferably, the predicted ride share experience information is related to the sequence in which the corresponding ride share order information is associated with the driver account.
In the technical scheme, the predicted carpooling experience information and the sequence of the carpooling order information associated to the driver account are further associated, and specifically, different carpooling experience information prediction standards are set according to the sequence of the information associated to the driver account. For example, the earlier the associated ride share order information, the more stringent the prediction criteria of the ride share experience information are to ensure the ride share experience of the accepted ride share order.
According to a second aspect of the embodiments of the present disclosure, there is provided a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the steps of the car pooling information processing method according to any one of the above technical solutions when executing the computer program, so that the method has all the beneficial technical effects of the car pooling information processing method, and details are not described herein.
According to a third aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the car pooling information processing method according to any of the above technical solutions, so that the method has all the beneficial technical effects of the car pooling information processing method, and is not described herein again.
Additional aspects and advantages of the disclosed embodiments will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the disclosed embodiments.
Drawings
The above and/or additional aspects and advantages of the embodiments of the present disclosure will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 shows a schematic flow diagram of a carpool information processing method according to one embodiment of the disclosure;
FIG. 2 shows a schematic flow diagram of a carpool information processing method according to another embodiment of the disclosure;
FIG. 3 shows a schematic flow diagram of a carpool information processing method according to yet another embodiment of the present disclosure;
FIG. 4 shows a schematic flow diagram of a carpool information processing method according to yet another embodiment of the present disclosure;
FIG. 5 shows an offline training flow diagram according to an embodiment one of the present disclosure;
FIG. 6 shows a flowchart of an online application according to a first embodiment of the present disclosure;
FIG. 7 shows an offline training flow diagram according to example two of the present disclosure;
FIG. 8 shows a flowchart of an online application according to embodiment two of the present disclosure;
FIG. 9 shows an offline training flow diagram according to example three of the present disclosure;
fig. 10 shows a flowchart of an online application according to a third embodiment of the present disclosure;
FIG. 11 shows a schematic flow chart diagram of a carpool information processing method according to yet another embodiment of the present disclosure;
FIG. 12 shows a schematic structural diagram of a computer device according to one embodiment of the present disclosure.
Detailed Description
In order that the above objects, features and advantages of the embodiments of the present disclosure can be more clearly understood, embodiments of the present disclosure will be described in further detail below with reference to the accompanying drawings and detailed description. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure, however, the embodiments of the disclosure may be practiced in other ways than those described herein, and therefore the scope of the embodiments of the disclosure is not limited by the specific embodiments disclosed below.
An embodiment of a first aspect of the disclosed embodiments provides a car pooling information processing method.
Fig. 1 shows a schematic flow diagram of a carpool information processing method according to one embodiment of the disclosure.
As shown in fig. 1, the car pooling information processing method includes:
s102, counting the incidence relation among historical carpool order information, historical user portrait information and historical carpool experience information, wherein the historical carpool experience information comprises forward data and backward data;
s104, receiving carpool order information;
and S106, predicting the carpooling experience information corresponding to the carpooling order information according to the carpooling order information, the zeroed user portrait information and the association relation.
According to the car sharing information processing method provided by the embodiment of the disclosure, the incidence relation among the historical car sharing order information, the historical user portrait information and the historical car sharing experience information in the historical data is counted, and on the basis of predicting whether the car sharing is right-way or not based on the objective car sharing order information, the user portrait information is introduced to eliminate the bias of the car sharing experience information on different individuals (people) possibly brought by subjective characteristics, so that the weights of different data in the historical car sharing order information are more reasonable when the incidence relation is counted, and the prediction accuracy is improved. However, it has been found that directly using the correlation to predict whether the carpool is going along the way can greatly increase the accuracy (up to ninety percent or more), but can lead to killing. For example, a user's tolerance is high, which results in the carpool assigned to the user being farther and farther, while a user's tolerance is low, which results in the carpool being better and better. When the incidence relation is used, the user portrait information is set to be zero only by adopting the car pooling order information, and the user portrait information is abandoned, so that the killing can be avoided, and the killing can be avoided while the prediction accuracy is improved. In addition, the user portrait information is set to zero when the incidence relation is used, the data processing amount during calculation can be reduced, and the hardware operation load is reduced, so that the hardware requirement for predicting the carpooling experience by using the incidence relation can be reduced, the calculation efficiency can be improved, and the calculation time is shortened. It can be understood that, like the historical carpooling experience information, the carpooling experience information also includes the on-road data and the off-road data; in addition, the carpooling experience information predicted by the carpooling information processing method can be used as a reference for whether filtered orders are recalled or not, and can also be used as a reference for whether carpooling is carried out on a plurality of orders or not.
In some embodiments, the step of predicting the car pooling experience information corresponding to the car pooling order information according to the car pooling order information, the zeroed user portrait information and the association relationship comprises: determining the degree of following corresponding to the carpool order information according to the carpool order information, the zeroed user portrait information and the association relation; and predicting the carpooling experience information corresponding to the carpooling order information according to the road degree.
In this embodiment, how to apply the statistically derived incidence relation to predict the ride share experience information is specifically defined. The method comprises the steps of firstly substituting car pooling order information and zeroed user portrait information into an incidence relation to obtain the following degree of the car pooling order information so as to reflect the following degree of the order, then predicting car pooling experience information according to the following degree, quantifying the following degree and improving the prediction precision.
Fig. 2 shows a schematic flow diagram of a carpool information processing method according to another embodiment of the disclosure.
As shown in fig. 2, the car pooling information processing method includes:
s202, counting the incidence relation among historical carpool order information, historical user portrait information and historical carpool experience information, wherein the historical carpool experience information comprises forward data and backward data;
S204, receiving carpool order information;
s206, determining the degree of following corresponding to the carpool order information according to the carpool order information, the zeroed user portrait information and the association relation;
s208, judging whether the road degree is more than or equal to a road degree threshold value, if so, turning to S210, and if not, turning to S212;
s210, predicting the carpooling experience information corresponding to the carpooling order information as the forward data;
s212, the carpooling experience information corresponding to the carpooling order information is predicted to be the out-of-way data.
In this embodiment, how to predict ride share experience information according to the degree of following the road is specifically defined. The forward degree threshold value is set to represent the forward degree during forward road, when the forward degree is larger than or equal to the forward degree threshold value, the carpooling experience information can be considered as forward data, otherwise, when the forward degree is smaller than the forward degree threshold value, the carpooling experience information is considered as out-of-road data, on one hand, clear judgment standards can be provided, on the other hand, the judgment standards during prediction can be changed by adjusting the forward degree threshold value, the prediction flexibility is improved, and the method is suitable for various scenes.
In some embodiments, the car pooling information processing method further comprises: and determining a forward degree threshold according to the accuracy of the historical predicted forward degree determined by the application incidence relation.
In this embodiment, it is further defined that the ride share information processing method further comprises how to determine a degree of compliance threshold. After the incidence relation obtained by statistics is applied to the carpooling order information, the corresponding road following degree can be predicted, and when the order corresponding to the carpooling order information is completed, the actual road following condition of the order can be obtained, so that whether the prediction result is accurate or not can be known. After a certain amount of application history data obtained by applying the incidence relation is accumulated, the accuracy of historical road degree prediction can be counted, and a reasonable road degree threshold value is determined according to the accuracy, so that the prediction accuracy is improved.
Specifically, the forward degree is a series of scores, each of the carpool orders in the application history data is classified according to the value of the historical prediction forward degree, the total quantity of the carpool orders corresponding to each historical prediction forward degree can be obtained, then the quantity of the carpool orders actually in the historical prediction forward degree is counted, and the proportion of the actual forward order in the historical prediction forward degree to all the orders (namely the ratio of the quantity of the actual forward carpool orders to the total quantity of the carpool orders corresponding to the historical prediction forward degree) is the accuracy of the historical prediction forward degree. Theoretically, the greater the calculated road following degree is, the more likely the road following is actually, and the smaller the value of the road following degree is but the lower the possibility of high accuracy is, so that a historical predicted road following degree with relatively high accuracy can be selected as a threshold value of the road following degree, the standard of predicting the carpooling experience information as the road following data can be improved, and the carpooling experience of the user is ensured. It is contemplated that the degree of compliance threshold may also continue to be adjusted in due time as application history data gradually accumulates.
In addition, when the predicted carpooling experience information is used as a reference for recalling the filtered orders, the concept of recall rate is also involved, the recall rate refers to the proportion of the recalled orders in the same degree of smoothness to the total amount of the filtered orders, for example, in the filtered orders, the number of orders with the degree of smoothness of 0.7 is 100, but only 50 of the orders are recalled, and the recall rate with the degree of smoothness of 0.7 is 50%. When the threshold value of the degree of the way ahead is determined, only the accuracy is used without considering the recall rate, namely the recall is lost, the accuracy is preferentially ensured, and the carpooling experience can be ensured not to be worsened.
Fig. 3 shows a schematic flow chart of a carpool information processing method according to yet another embodiment of the present disclosure.
As shown in fig. 3, the car pooling information processing method includes:
s302, carrying out statistics on the incidence relation among historical car pooling order information, historical user portrait information, historical car pooling experience information and the following road degree by using a deep learning algorithm, wherein the historical car pooling experience information comprises following road data and non-following road data;
s304, receiving carpool order information;
s306, determining the degree of following corresponding to the carpool order information according to the carpool order information, the zeroed user portrait information and the association relation;
And S308, predicting the carpooling experience information corresponding to the carpooling order information according to the road degree.
In this embodiment, a scheme of statistical association is specifically defined. The association relationship is counted by using a deep learning algorithm, that is, a prediction model (which can be a general model) is firstly created, and the model is trained by using historical data, so that the prediction model capable of reflecting the association relationship can be obtained. Specifically, when the model is trained, the historical car pooling order information and the historical user portrait information are used as training parameters, the historical car pooling experience information is used as a mark, the degree of the way is used as a target parameter, and training is performed, so that the prediction model can naturally embody the reasonable weight relationship between the car pooling order information and the user portrait information, and the model training process is facilitated to be simplified. When the model is applied, the carpool order information is input into the model, the user image information in the model is set to zero, namely the weight is removed, and the prejudice possibly brought by the main factor is eliminated by the output degree of the way.
Fig. 4 shows a schematic flow diagram of a carpool information processing method according to yet another embodiment of the present disclosure.
As shown in fig. 4, the car pooling information processing method includes:
S402, establishing a first incidence relation among historical carpool order information, historical carpool experience information and a first road following degree, wherein the historical carpool experience information comprises road following data and road non-following data;
s404, creating a second association relation among historical user portrait information, historical carpooling experience information and a second road degree;
s406, performing joint statistics on the first association relation and the second association relation by using the same historical car pooling order information, historical user portrait information and historical car pooling experience information;
s408, receiving carpool order information;
s410, determining a first road degree according to the car pooling order information and the first association relation, and taking the first road degree as the road degree corresponding to the car pooling order information;
and S412, predicting the carpooling experience information corresponding to the carpooling order information according to the road degree.
In this embodiment, another scheme of the statistical association relationship is specifically defined. Respectively establishing an association relation with the degree of following the road aiming at the historical carpool order information and the historical user portrait information, namely a first association relation and a second association relation, performing combined statistics on the first association relation and the second association relation by using the same historical data during statistics to associate the first association relation and the second association relation, wherein the statistical results can respectively reflect the influence of two factors on the degree of following the road, and the differentiation and association are realized.
When the statistical result is used for predicting the road following degree, the scheme of the embodiment of the disclosure is that the user portrait information is abandoned, so that the car pooling order information can be directly substituted into the first association relationship, and the output first road following degree not only reflects the influence of the user portrait information, but also is a result of prediction only by using the car pooling order information but not the user portrait information, so that the prediction accuracy can be improved and the killing can be avoided. In addition, because only the first association relation is used, only the first association relation can be stored and operated on hardware, and the storage and calculation load is reduced.
It is conceivable that, in this scheme, a deep learning algorithm may also be adopted to respectively establish a first prediction model and a second prediction model, and specifically, both the models may be modeled by Logistic Regression, the training parameters of the first prediction model are historical carpool order information and are marked as historical carpool experience information, the target parameters are first road degrees, the training parameters of the second prediction model are historical user portrait information and are marked as historical carpool experience information, and the target parameters are second road degrees, and then the two models are subjected to combined training, so that two prediction models which are distinguished and associated can be obtained. When the model is applied, if only one of the car pooling order information and the user portrait information is used, for example, only the car pooling order information is used in the embodiment of the disclosure, the information to be used is input into the corresponding model, and the output road degree is taken as the predicted road degree; and if the carpool order information and the user portrait information need to be used simultaneously, respectively inputting the carpool order information and the user portrait information into the two models, and integrating the obtained first road degree and the second road degree into a road degree according to a joint mode during joint training.
In some embodiments, the joint statistics of the first association relation and the second association relation includes: and performing joint statistics on the first association relation and the second association relation in an addition or multiplication mode.
In this embodiment, a manner of performing joint statistics on the first association relationship and the second association relationship is specifically defined. The method can adopt an addition mode, namely weights are respectively given to the first road degree and the second road degree during statistics, so that the weighted value of the first road degree and the second road degree is 1; it is also possible to use a multiplication method, i.e. to pre-multiply the two output results first and then normalize them during statistics. Both of these ways can achieve joint statistics.
In some embodiments, the ride share order information includes at least one of: the method comprises the steps of ordering original time length, ordering original length, ordering estimated time length, ordering estimated length, estimated detour distance, estimated detour time, estimated driving receiving time, estimated departure time and travel scene; and/or the historical carpool order information includes at least one of: the order original time length, the order original length, the order car sharing time length, the order car sharing length, the detour distance, the detour time, the driving receiving time, the departure time and the travel scene.
In this embodiment, the order sharing information and the data included in the historical order sharing information are specifically defined, and these data can be regarded as the characteristics of the order dimension. The method comprises the steps that the order original time length, the order original length and the trip scene can be determined when orders are placed, other characteristics can be determined only after corresponding events occur, for example, the order car-sharing time length, the order car-sharing length, the detour distance and the detour time can be determined only when orders are completed, the pickup time can be determined only when pickup is successful, the departure time can be determined only when the order is departed, and when the characteristics correspond to current car-sharing order information to be predicted, the order estimated time length, the estimated detour distance, the estimated detour time, the estimated pickup time and the estimated departure time are correspondingly used for replacing so as to predict car-sharing experience information. By introducing the characteristics, the situation of the current carpool order can be fully known so as to predict carpool experience information. It should be noted that, the travel scene is related to the destination, time, and other factors, and may be mined by a special algorithm, which is not described in detail herein, for example, the travel scene may be an early peak scene, a work scene, a travel scene, and the like, and the travel requirements under different scenes are different, and the degree of forward road prediction may be adjusted accordingly to improve the prediction flexibility and the adaptability. For example, the travel pressure in the early peak scene is high, and the user can be satisfied as long as the user can get on the vehicle, so the requirement on the forward road is relatively low, and the output forward road degree can be relatively higher for the same starting point and the same finishing point relative to other scenes, so that the spelling rate is improved; under the travel scene, the destination is a tourist attraction, the requirement on time is lower at the moment, but the requirement on the forward road is higher, the output forward road degree can be lower than that of other scenes, and the carpooling quality is improved.
In some embodiments, the user representation information includes at least one of: gender, age, occupation, monthly outstanding passenger inventory, weekly driver inventory, driver rating, historical passenger feedback inventory, driver cancellation rate, driver response rate, and passenger cancellation rate.
In this embodiment, data contained in user portrait information is specifically defined, which may be considered portrait features, where users refer to passengers and drivers. By introducing the characteristics, the subjective emotion of the user can be fully reflected, and the accuracy of prediction is improved. In consideration of the fact that the frequency of orders sent by passengers is far lower than the frequency of orders received by a driver, the monthly finished-order quantity of the passengers and the weekly finished-order quantity of the drivers are respectively used in the finished-order quantity so as to ensure the balance of the collected data. It can be understood that, unlike some of the data in the car pool order information which is estimated, the data of the user portrait information is clear data, so the data included in the historical user portrait information and the user portrait information are the same.
Three schemes of statistics and application of association relations are introduced in the following three embodiments, wherein the association relations are all embodied in the form of models, historical portrait features of passengers and users and historical order dimension features are used in the model training, the order dimension features are used in the model application, and the following three schemes respectively correspond to three model training modes.
Example one
1. Performing on-line training: as shown in fig. 5, the historical portrait features and the historical order dimension features of the passenger and the driver are obtained, the historical portrait features and the historical order dimension features of the passenger and the driver are input into the same model (which may be a general model) for training, the model is marked as whether to be on the way or not during training, and the training objective function is the degree of the way (between 0 and 1). In the training process, based on the objective function generation rule in the model, there may be an offline threshold (e.g. 0.5), if an order sample itself is not in-route, the output in-route degree will be smaller than the offline threshold, otherwise, if an order sample itself is in-route, the output in-route degree will be greater than or equal to the offline threshold, and the value of the in-route degree is larger and the more the in-route degree is.
2. The on-line application comprises the following steps: as shown in fig. 6, the characteristics of the order dimension are obtained and input into the model. The model sets the portrait characteristics of passengers and drivers to zero, which is equal to taking the weight away, then outputs the forward road degree, judges whether the forward road degree is larger than or equal to the threshold value on the line, if yes, the forward road is followed, if not, the forward road is not followed, and the forward road is followed the larger the forward road degree value is, and outputs the forward road or the non-forward road.
In fact, there are often a plurality of order samples carried by one driver account, and there may be a training model in which a part of samples carried by the same driver account are on-way and another part are off-way. The on-line threshold is determined by selecting a higher accuracy on-line degree for recall according to the accuracy and recall rate of the model under each on-line degree score (the recall is lost and the accuracy is preferentially ensured because the on-line experience is ensured not to be worsened).
Example two
1. Performing on-line training: as shown in FIG. 7, one model1 was created using the characteristics of historical order dimensions, and another model2 was created using the characteristics of historical figures of passengers and drivers, both models being modeled with Logistic Regression. Then, a weight w1 is given to the prediction result of the model1, a weight w2 is given to the prediction result of the model2, the constraint condition w1+ w2 is met, and finally the two models are subjected to joint training, namely, the training mode is model addition. During training, the label of model modeling is still whether the road is right, the training objective function is still right, the training process is similar to the first embodiment, and details are not repeated here.
2. The on-line application comprises the following steps: as shown in fig. 8, the characteristics of the order dimension are obtained, the characteristics of the order dimension are input into a trained model1, and the model1 outputs the prediction cis-degree. And judging whether the road is direct or not according to the on-line threshold value, and outputting the direct or non-direct. See embodiment one for the on-line threshold.
EXAMPLE III
1. Performing on-line training: as shown in FIG. 9, one model3 was created using the characteristics of historical order dimensions, and another model4 was created using the characteristics of historical figures of passengers and drivers, both models being modeled with Logistic Regression. Then, the prediction result of the model3 and the prediction result of the model4 are pre-multiplied and then normalized (mainly to pull the degree of the road back to the normal interval of 0 to 1), and the two models are jointly trained in a multiplication mode. During training, the label of model modeling is still whether the road is right, the training objective function is still right, the training process is similar to the first embodiment, and details are not repeated here.
Specifically, for example, if the historical order feature model3 predicts a first degree of compliance of 0.6 and the historical image feature model4 predicts a second degree of compliance of 0.7, then the output degree of compliance is 0.6 × 0.7 — 0.42 and the degree of non-compliance is 0.4 × 0.3 — 0.12, and then the two scores (0.42 and 0.12) are normalized and the model biases cancel.
2. The on-line application comprises the following steps: as shown in fig. 10, the characteristics of the order dimensions are acquired, the characteristics of the order dimensions are input to a trained model3, the predicted forward road degree is output, whether the road is forward or not is judged according to the on-line threshold, and the forward road or the non-forward road is output. See embodiment one for the on-line threshold.
In summary, the embodiment of the present disclosure provides an order-based ride share experience assessment method, which eliminates prejudice of ride share on different individuals (people) possibly brought by subjective features by introducing user portrait features of passengers and drivers in a model based on an original order-feature-based ride share estimation model, so that weights of different features of learned order dimensions are more reasonable. By adjusting the training mode, the accuracy is improved, but the portrait characteristics of passengers and drivers are abandoned when the model is used, and the killing is avoided. Three different training modes can respectively improve the accuracy of model prediction by one percent, three percent and four percent.
Fig. 11 shows a schematic flow chart of a carpool information processing method according to still another embodiment of the present disclosure.
As shown in fig. 11, the car pool information processing method includes:
S502, counting the incidence relation among historical car pooling order information, historical user portrait information and historical car pooling experience information, wherein the historical car pooling experience information comprises forward data and backward data;
s504, receiving at least two pieces of carpooling order information of the carpooling to be performed;
s506, predicting the carpooling experience information corresponding to the carpooling order information according to the carpooling order information, the zeroed user portrait information and the association relation;
and S508, determining that the carpooling experience information corresponding to all the carpooling order information to be carpooled is the forward data, and associating all the carpooling order information to the same driver account.
In this embodiment, a scheme of completing car sharing by using the prediction result is specifically defined. The background can receive the carpooling requests sent by different passenger users, which can be newly sent by the passengers or filtered carpooling requests, the requests include basic planned starting time and starting and ending points, when the carpooling operation is executed or whether the filtered carpooling requests need to be recalled, whether a plurality of carpooling requests to be carpooled are in-route or not needs to be calculated, at the moment, a driving route can be planned based on the carpooling requests, carpooling order information corresponding to the carpooling requests is generated, carpooling experience information is respectively predicted for each piece of carpooling order information, when the predicted carpooling experience information is all in-route data, the carpooling requests are determined or recalled, carpooling is carried out, and the carpooling order information of the determined carpooling is related to the same account number, and the carpooling operation is completed. The scheme can ensure the carpooling experience of the user.
It is conceivable that, when generating the car pooling order information, the real-time position corresponding to the driver account and the car pooling order situation carried by the driver account can be combined, so that the route planning and prediction are more accurate. Further, when the driver account already carries the car-sharing order, the car-sharing order already carried by the driver account needs to be combined with the newly received car-sharing request, a new driving route is planned, new car-sharing order information is generated, the car-sharing experience information is predicted again according to each piece of car-sharing order information, and at the moment, the new car-sharing order is carried when all pieces of car-sharing experience information are still required to be guaranteed to be the in-route data.
In some embodiments, the predicted ride share experience information is related to the chronological order in which the corresponding ride share order information is associated with the driver account.
In this embodiment, the predicted carpooling experience information and the sequence of the carpooling order information associated with the driver account are further associated, specifically, different carpooling experience information prediction standards are set according to the sequence of the information associated with the driver account. For example, the earlier associated car-sharing order information may be, the stricter the prediction standard of the car-sharing experience information is, specifically, the higher the smooth road threshold value is, at this time, if a new car-sharing request is received, it is necessary to re-plan a route for the car-sharing order already carried by the driver account and the newly received car-sharing request, and generate new car-sharing order information and re-predict car-sharing experience information respectively, and if it is found that the car-sharing experience information of the car-sharing order already carried by the driver account becomes the non-smooth road data after the re-prediction, the new car-sharing request is not received, so as to ensure the car-sharing experience of the carried car-sharing order.
As shown in fig. 12, an embodiment of a second aspect of the embodiment of the present disclosure provides a computer device 1, which includes a memory 12, a processor 14, and a computer program stored on the memory 12 and capable of running on the processor 14, and when the processor 14 executes the computer program, the steps of the car sharing information processing method according to any one of the above embodiments are implemented, so that all beneficial technical effects of the car sharing information processing method are achieved, and details are not described herein.
In particular, the memory 12 described above may include mass storage for data or instructions. By way of example, and not limitation, memory 12 may include a Hard Disk Drive (HDD), a floppy Disk Drive, flash memory, an optical Disk, a magneto-optical Disk, tape, or a Universal Serial Bus (USB) Drive or a combination of two or more of these. Memory 12 may include removable or non-removable (or fixed) media, where appropriate. The memory 12 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 12 is a non-volatile solid-state memory. In a particular embodiment, the memory 12 includes Read Only Memory (ROM). Where appropriate, the ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), electrically rewritable ROM (EAROM), or flash memory or a combination of two or more of these.
Processor 14 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured as one or more integrated circuits implementing embodiments of the present disclosure.
An embodiment of the third aspect of the embodiments of the present disclosure provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the car pooling information processing method according to any of the embodiments, so that all the beneficial technical effects of the car pooling information processing method are achieved, and are not described herein again.
Computer readable storage media may include any medium that can store or transfer information. Examples of computer readable storage media include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the disclosed embodiments should be included in the scope of protection of the disclosed embodiments.
Claims (13)
1. A car pooling information processing method is characterized by comprising the following steps:
counting the incidence relation among historical carpool order information, historical user portrait information and historical carpool experience information, wherein the historical carpool experience information comprises forward data and backward data;
receiving carpooling order information;
and predicting the carpooling experience information corresponding to the carpooling order information according to the carpooling order information, the zeroed user portrait information and the incidence relation.
2. The car pooling information processing method according to claim 1, wherein the step of predicting car pooling experience information corresponding to the car pooling order information according to the car pooling order information, the zeroed user portrait information and the association relationship comprises:
determining the degree of a road corresponding to the carpool order information according to the carpool order information, the zeroed user portrait information and the incidence relation;
and predicting the carpooling experience information corresponding to the carpooling order information according to the road degree.
3. The method for processing the carpooling information according to claim 2, wherein the step of predicting the carpooling experience information corresponding to the carpooling order information according to the degree of the way comprises:
Determining that the degree of following is greater than or equal to a threshold value of the degree of following, and predicting that the carpooling experience information corresponding to the carpooling order information is following data;
and determining that the degree of following is smaller than the threshold value of the degree of following, and predicting that the carpooling experience information corresponding to the carpooling order information is out-of-way data.
4. The carpooling information processing method according to claim 3, further comprising:
and determining the degree of road threshold according to the accuracy of the historical predicted degree of road determined by applying the incidence relation.
5. The carpooling information processing method according to any one of claims 2 to 4, wherein the step of counting the association relationship between the historical carpooling order information, the historical user portrait information and the historical carpooling experience information comprises:
and counting the incidence relation among the historical car pooling order information, the historical user portrait information, the historical car pooling experience information and the road following degree by utilizing a deep learning algorithm.
6. The carpooling information processing method according to any one of claims 2 to 4, wherein the step of counting the association relationship between the historical carpooling order information, the historical user portrait information and the historical carpooling experience information comprises:
Creating a first incidence relation among the historical carpooling order information, the historical carpooling experience information and a first road degree;
creating a second incidence relation among the historical user portrait information, the historical carpooling experience information and a second road degree;
and performing joint statistics on the first association relation and the second association relation by using the same historical car pooling order information, the same historical user portrait information and the same historical car pooling experience information.
7. The car pooling information processing method according to claim 6, wherein the step of determining a degree of following corresponding to said car pooling order information according to said car pooling order information, said zeroed user portrait information and said association relationship comprises:
and determining the first road degree according to the car pooling order information and the first association relation, and taking the first road degree as the road degree corresponding to the car pooling order information.
8. The car pooling information processing method according to claim 6, wherein the operation of performing joint statistics on the first association relation and the second association relation comprises:
and performing joint statistics on the first incidence relation and the second incidence relation in an addition or multiplication mode.
9. The carpooling information processing method according to any one of claims 1 to 4,
the carpool order information comprises at least one of the following information: the method comprises the steps of ordering original time length, ordering original length, ordering estimated time length, ordering estimated length, estimated detour distance, estimated detour time, estimated driving receiving time, estimated departure time and travel scene; and/or
The historical carpooling order information comprises at least one of the following: the method comprises the steps of obtaining an original order time length, an original order length, an order carpooling time length, an order carpooling length, a detour distance, a detour time, a driving receiving time, a starting time and a travel scene; and/or
The user representation information includes at least one of: gender, age, occupation, monthly outstanding passenger inventory, weekly driver inventory, driver rating, historical passenger feedback inventory, driver cancellation rate, driver response rate, and passenger cancellation rate.
10. The carpooling information processing method according to any one of claims 1 to 4,
the step of receiving the carpool order information comprises the following steps:
receiving at least two pieces of car sharing order information to be shared;
the car pooling information processing method further comprises the following steps:
determining that the carpooling experience information corresponding to all the carpooling order information to be carpooled is the forward data, and associating all the carpooling order information to the same driver account.
11. The car pooling information processing method according to claim 10,
and the predicted carpooling experience information is related to the sequence of the carpooling order information related to the driver account.
12. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and being executable on the processor, characterized in that the processor realizes the steps of the ride share information processing method according to any of claims 1 to 11 when executing the computer program.
13. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the ride share information processing method according to any one of claims 1 to 11.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910959634.2A CN111861617A (en) | 2019-10-10 | 2019-10-10 | Car pooling information processing method, computer device and computer readable storage medium |
| PCT/CN2020/120219 WO2021068944A1 (en) | 2019-10-10 | 2020-10-10 | Carpool order processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910959634.2A CN111861617A (en) | 2019-10-10 | 2019-10-10 | Car pooling information processing method, computer device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111861617A true CN111861617A (en) | 2020-10-30 |
Family
ID=72970650
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910959634.2A Pending CN111861617A (en) | 2019-10-10 | 2019-10-10 | Car pooling information processing method, computer device and computer readable storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111861617A (en) |
| WO (1) | WO2021068944A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112330146A (en) * | 2020-11-04 | 2021-02-05 | 杭州拼便宜网络科技有限公司 | Virtual resource allocation method, device, equipment and readable storage medium |
| CN112561639A (en) * | 2020-12-15 | 2021-03-26 | 北京嘀嘀无限科技发展有限公司 | Car pooling service processing method, device, equipment and storage medium |
| CN117236473A (en) * | 2023-11-10 | 2023-12-15 | 华中科技大学 | Method and system for distributing paths of passenger co-riding vehicles in junction evacuation scene |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115727861A (en) * | 2021-08-25 | 2023-03-03 | 北京顺丰同城科技有限公司 | Vehicle path planning method and device, computer equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106503821A (en) * | 2016-11-01 | 2017-03-15 | 成都俊巡科技有限公司 | Realize the Carpooling system of car owner and the two-way invitation function of passenger |
| CN108573315A (en) * | 2018-04-09 | 2018-09-25 | 北京嘀嘀无限科技发展有限公司 | A kind of prompt message determines method, system and computer readable storage medium |
| CN109478275A (en) * | 2017-06-16 | 2019-03-15 | 北京嘀嘀无限科技发展有限公司 | The system and method for distributing service request |
| JP2019096263A (en) * | 2017-11-28 | 2019-06-20 | 日産自動車株式会社 | Information providing method and information providing device |
| CN110245953A (en) * | 2019-05-20 | 2019-09-17 | 深圳市轱辘汽车维修技术有限公司 | A kind of Information Authentication method, Information Authentication device and electronic equipment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9904900B2 (en) * | 2015-06-11 | 2018-02-27 | Bao Tran | Systems and methods for on-demand transportation |
| CN106709591A (en) * | 2016-08-11 | 2017-05-24 | 淮阴工学院 | Cooperative car-pooling route selection method with uncertain demand in Internet of Vehicles environment |
| EP3623764A1 (en) * | 2017-01-25 | 2020-03-18 | Via Transportation, Inc. | Method and system for managing a fleet of ride-sharing vehicles using virtual bus stops |
| CN108734361B (en) * | 2017-04-18 | 2021-12-03 | 北京嘀嘀无限科技发展有限公司 | Car pooling order processing method and device |
-
2019
- 2019-10-10 CN CN201910959634.2A patent/CN111861617A/en active Pending
-
2020
- 2020-10-10 WO PCT/CN2020/120219 patent/WO2021068944A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106503821A (en) * | 2016-11-01 | 2017-03-15 | 成都俊巡科技有限公司 | Realize the Carpooling system of car owner and the two-way invitation function of passenger |
| CN109478275A (en) * | 2017-06-16 | 2019-03-15 | 北京嘀嘀无限科技发展有限公司 | The system and method for distributing service request |
| JP2019096263A (en) * | 2017-11-28 | 2019-06-20 | 日産自動車株式会社 | Information providing method and information providing device |
| CN108573315A (en) * | 2018-04-09 | 2018-09-25 | 北京嘀嘀无限科技发展有限公司 | A kind of prompt message determines method, system and computer readable storage medium |
| CN110245953A (en) * | 2019-05-20 | 2019-09-17 | 深圳市轱辘汽车维修技术有限公司 | A kind of Information Authentication method, Information Authentication device and electronic equipment |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112330146A (en) * | 2020-11-04 | 2021-02-05 | 杭州拼便宜网络科技有限公司 | Virtual resource allocation method, device, equipment and readable storage medium |
| CN112330146B (en) * | 2020-11-04 | 2024-01-23 | 杭州拼便宜网络科技有限公司 | Virtual resource allocation method, device, equipment and readable storage medium |
| CN112561639A (en) * | 2020-12-15 | 2021-03-26 | 北京嘀嘀无限科技发展有限公司 | Car pooling service processing method, device, equipment and storage medium |
| CN117236473A (en) * | 2023-11-10 | 2023-12-15 | 华中科技大学 | Method and system for distributing paths of passenger co-riding vehicles in junction evacuation scene |
| CN117236473B (en) * | 2023-11-10 | 2024-01-30 | 华中科技大学 | Method and system for distributing paths of passenger co-riding vehicles in junction evacuation scene |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021068944A1 (en) | 2021-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111861617A (en) | Car pooling information processing method, computer device and computer readable storage medium | |
| US8229666B2 (en) | Generating and using pattern keys in navigation systems to predict user destinations | |
| US8180557B2 (en) | Traffic state predicting apparatus | |
| WO2019212600A1 (en) | Deep reinforcement learning for optimizing carpooling policies | |
| CN111325374B (en) | Method and device for predicting order cancellation probability and electronic equipment | |
| CN108444486B (en) | Navigation route sorting method and device | |
| CN110118567B (en) | Travel mode recommendation method and device | |
| US20170243172A1 (en) | Cognitive optimal and compatible grouping of users for carpooling | |
| US20200210905A1 (en) | Systems and Methods for Managing Networked Vehicle Resources | |
| JP5070574B2 (en) | Local traffic prediction program generation device, local traffic prediction device, local traffic prediction program generation method, local traffic prediction method and program | |
| JP2018173697A (en) | Shared use charge calculation system | |
| CN112650825A (en) | Method and device for determining abnormal drive receiving behavior, storage medium and electronic equipment | |
| US10949751B2 (en) | Optimization of multiple criteria in journey planning | |
| CN111862583B (en) | Traffic flow prediction method and device | |
| CN111737601A (en) | Method, device and equipment for recommending travel strategy and storage medium | |
| CN112985442B (en) | Driving path matching method, readable storage medium and electronic device | |
| US20190340315A1 (en) | Method and device for providing vehicle navigation simulation environment | |
| CN113418531A (en) | Navigation route determination method and device, electronic equipment and computer storage medium | |
| JP2021524574A (en) | Devices and methods for outputting navigation information and automobiles | |
| CN110874777A (en) | Order processing method and device | |
| CN113902209A (en) | Travel route recommendation method, edge server, cloud server, equipment and medium | |
| US20190339086A1 (en) | Method and device for providing vehicle navigation simulation environment | |
| CN117278940B (en) | Car machine interconnection system based on hicar | |
| CN116358582A (en) | Training method, device and medium for recommended and acquainted road prediction model of navigation route | |
| CN112734473B (en) | Network appointment vehicle blockage-avoiding detour identification method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |