Generation of trip estimates using real-time data and historical data
A system uses machine models to estimate trip durations or distance. The system trains a historical model to estimate trip duration using characteristics of past trips. The system trains a real-time model to estimate trip duration using characteristics of recently completed trips. The historical and real-time models may use different time windows of training data to predict estimates, and may be trained to predict an adjustment to an initial trip estimate. A selector model is trained to predict whether the historical model, the real-time model, or a combination of the historical and real-time models will more accurately estimate a trip duration, given features associated with a trip duration request, and the system accordingly uses the models to estimate a trip duration. In some embodiments, the real-time model and the selector may be trained using batch machine learning techniques which allow the models to incorporate new trip data as trips complete.
1. A computer-implemented method comprising:
calculating, by a computer processor, a first set of confidence intervals representing estimated durations by applying a real-time computer model to a set of inputs;
calculating, by the computer processor, a second set of confidence intervals representing estimated durations by applying a historical computer model to a set of inputs; and
determining, by the computer processor, a final confidence interval by applying a selector computer model which determines whether the real-time computer model or the historical computer model is likely to more accurately predict durations to the first set of confidence intervals and the second set of confidence intervals.
2. The method of claim 1 , further comprising:
predicting an initial set of durations based on a network-based estimation algorithm;
wherein the real-time computer model and the historical computer model use the predicted set of initial durations to generate the first set of confidence intervals and the second set of confidence intervals.
3. The method of claim 2 , wherein the network-based estimation algorithm bases duration predictions on a graph that models a path between an origin location of a trip and a destination location of the trip as a series of weighted nodes and edges.
4. The method of claim 1 , wherein upper and lower bounds of a confidence interval correspond to a quartile values.
5. The method of claim 4 , wherein the quartile values are assigned weighting values according to their accuracy and used to train a model to determine the final confidence interval.
6. The method of claim 1 , wherein the real-time computer model is trained using features of trips that completed within a first time period, and the historical computer model is trained using features of trips that completed within a second time period.
7. A non-transitory computer-readable storage medium storing computer program instructions executable by one or more processors of a system to perform steps comprising:
calculating, by the one or more computer processors, a first set of confidence intervals representing estimated durations by applying a real-time computer model to a set of inputs;
calculating, by the one or more computer processors, a second set of confidence intervals representing estimated durations by applying a historical computer model to a set of inputs; and
determining, by the one or more computer processors, a final confidence interval by applying a selector computer model which determines whether the real-time computer model or the historical computer model is likely to more accurately predict durations to the first set of confidence intervals and the second set of confidence intervals.
8. The non-transitory computer-readable storage medium of claim 7 , further comprising:
predicting an initial set of durations based on a network-based estimation algorithm;
wherein the real-time computer model and the historical computer model use the predicted set of initial durations to generate the first set of confidence intervals and the second set of confidence intervals.
9. The non-transitory computer-readable storage medium of claim 8 , wherein the network-based estimation algorithm bases duration predictions on a graph that models a path between an origin location of a trip and a destination location of the trip as a series of weighted nodes and edges.
10. The non-transitory computer-readable storage medium of claim 7 , wherein upper and lower bounds of a confidence interval correspond to a quartile values.
11. The non-transitory computer-readable storage medium of claim 10 , wherein the quartile values are assigned weighting values according to their accuracy and used to train a model to determine the final confidence interval.
12. The non-transitory computer-readable storage medium of claim 7 , wherein the real-time computer model is trained using features of trips that completed within a first time period, and the historical computer model is trained using features of trips that completed within a second time period.
13. A computer system comprising:
one or more computer processors for executing computer program instructions; and
a non-transitory computer-readable storage medium storing instructions executable by the one or more computer processors to perform steps comprising:
calculating, by the one or more computer processors, a first set of confidence intervals representing estimated durations by applying a real-time computer model to a set of inputs;
calculating, by the one or more computer processors, a second set of confidence intervals representing estimated durations by applying a historical computer model to a set of inputs; and
determining, by the one or more computer processors, a final confidence interval by applying a selector computer model which determines whether the real-time computer model or the historical computer model is likely to more accurately predict durations to the first set of confidence intervals and the second set of confidence intervals.
14. The system of claim 13 , further comprising:
predicting an initial set of durations based on a network-based estimation algorithm;
wherein the real-time computer model and the historical computer model use the predicted set of initial durations to generate the first set of confidence intervals and the second set of confidence intervals.
15. The system of claim 14 , wherein the network-based estimation algorithm bases duration predictions on a graph that models a path between an origin location of a trip and a destination location of the trip as a series of weighted nodes and edges.
16. The system of claim 13 , wherein upper and lower bounds of a confidence interval correspond to a quartile values.
17. The system of claim 16 , wherein the quartile values are assigned weighting values according to their accuracy and used to train a model to determine the final confidence interval.