Systems and methods for determining estimated time of arrival

1. A system for determining an estimated time of arrival (ETA), comprising:

a storage device storing a set of instructions; and

at least one processor of an online on-demand service platform configured to communicate with the storage device, wherein when executing the set of instructions, the at least one processor is configured to cause the system to:

receive a start location and a destination from a user device via a network;

obtain a machine learning model for determining an ETA, wherein the machine learning model is generated according to a process for generating a machine learning model, the process including:

obtaining historical data related to an on-demand service order,

determining a high-dimensional sparse feature based on the historical data, wherein the high-dimensional sparse feature includes at least one feature corresponding to distance information of a route, and

determining a machine learning model based on the high-dimensional sparse feature;

determine an ETA for a target route based on the machine learning model, the start location, and the destination; and

transmit, via the network, the determined ETA to the user device.

2. The system of claim 1 , wherein to determine the high-dimensional sparse feature involves at least one of a gradient boosting decision tree (GBDT) algorithm or a one-hot encoding algorithm.

3. The system of claim 1 , wherein to determine the high-dimensional sparse feature, the at least one processor is further configured to cause the system to:

obtain a feature from the historical data;

determine tree information based on the feature;

determine a leaf node encoding based on the tree information; and

determine the high-dimensional sparse feature based on the leaf node encoding.

4. The system of claim 1 , wherein to determine the high-dimensional sparse feature, the at least one processor is further configured to cause the system to:

obtain a feature from the historical data;

determine whether the feature is a categorical feature;

upon the feature being the categorical feature, determine an attribution corresponding to the feature;

determine a one-hot encoding based on the attribution; and

determine the high-dimensional sparse feature based on the one-hot encoding.

5. The system of claim 4 , wherein to determine the high-dimensional sparse feature, the at least one processor is further configured to cause the system to:

upon the feature being not the categorical feature, determine tree information based on the feature;

determine a leaf node encoding based on the tree information; and

determine the high-dimensional sparse feature based on the leaf node encoding and the one-hot encoding.

6. The system of claim 1 , wherein to determine the machine learning model based on the high-dimensional sparse feature involves at least one of factorization machine (FM) or field-aware factorization machine (FFM).

7. The system of claim 1 , wherein to obtain the machine learning model, the at least one processor is further configured to cause the system to update the machine learning model with an on-line learning method.

8. The system of claim 7 , wherein the on-line learning method involves at least of follow the regularized leader (FTRL), online gradient descent (OGD), stochastic gradient descent (SGD), forward-backward splitting method (FOBOS), or regularized dual averaging (RDA).

9. The system of claim 7 , wherein to update the machine learning model with the on-line learning method, the at least one processor is further configured to cause the system to:

obtain recent historical data related to another on-demand service order;

determine a recent high-dimensional sparse feature based on the recent historical data; and

update the machine learning model based on the recent high-dimensional sparse feature.

10. The system of claim 1 , wherein the historical data includes at least one of basic route information, vehicle information, or provider information.

11. A method for determining an ETA, comprising:

receiving, by a server via a network, a start location and a destination from a user device;

obtaining, by the server, a machine learning model for determining an ETA, wherein the machine learning model is generated according to a process, the process including:

obtaining historical data related to an on-demand service order,

determining a high-dimensional sparse feature based on the historical data, wherein the high-dimensional sparse feature includes at least one feature corresponding to distance information of a route, and

determining a machine learning model based on the high-dimensional sparse feature;

determining, by the server, an ETA for a target route based on the machine learning model, the start location, and the destination; and

transmitting, by the server via the network, the determined ETA to the user device.

12. The method of claim 11 , wherein the determining of the high-dimensional sparse feature involves at least one of a GBDT algorithm or a one-hot encoding algorithm.

13. The method of claim 11 , wherein the determining of the high-dimensional sparse feature further includes:

obtaining a feature from the historical data;

determining tree information based on the feature;

determining a leaf node encoding based on the tree information; and

determining the high-dimensional sparse feature based on the leaf node encoding.

14. The method of claim 11 , wherein the determining of the high-dimensional sparse feature further includes:

obtaining a feature from the historical data;

determining whether the feature is a categorical feature;

upon the feature being the categorical feature, determining an attribution corresponding to the feature;

determining a one-hot encoding based on the attribution; and

determining the high-dimensional sparse feature based on the one-hot encoding.

15. The method of claim 14 , wherein the determining of the high-dimensional sparse feature further includes:

upon the feature being not the categorical feature, determining tree information based on the feature;

determining a leaf node encoding based on the tree information; and

determining the high-dimensional sparse feature based on the leaf node encoding and the one-hot encoding.

16. The method of claim 11 , wherein the determining of the machine learning model based on the high-dimensional sparse feature involves at least one of FM or FFM.

17. The method of claim 11 , wherein the obtaining of the machine learning model further includes:

updating, by the server, the machine learning model with an on-line learning method.

18. The method of claim 17 , wherein the on-line learning method involves at least one of FTRL, OGD, SGD, FOBOS, or RDA.

19. The method of claim 17 , wherein the updating of the machine learning model with the on-line learning method further includes:

obtaining recent historical data related to another on-demand service order;

determining a recent high-dimensional sparse feature based on the recent historical data; and

updating the machine learning model based on the recent high- dimensional sparse feature.

20. A non-transitory machine-readable storage medium including instructions that, when accessed by at least one processor of an online on-demand service platform, causes the at least one processor to:

receive a start location and a destination from a user device via a network;

obtain a machine learning model for determining an ETA, wherein the machine learning model is generated according to a process for generating a machine learning model, the process including:

obtaining historical data related to an on-demand service order,

determining a high-dimensional sparse feature based on the historical data, wherein the high-dimensional sparse feature includes at least one feature corresponding to distance information of a route, and

determining a machine learning model based on the high-dimensional sparse feature;

determine an ETA for a target route based on the machine learning model, the start location, and the destination; and

transmit, via the network, the determined ETA to the user device.