Abstract
Amazon Last Mile strives to learn an accurate delivery point for each address by using the noisy GPS locations reported from past deliveries. Centroids and other center-finding methods do not serve well, because the noise is consistently biased. The problem calls for supervised machine learning, but how? We addressed it with a novel adaptation of learning to rank from the information retrieval domain. This also enabled information fusion from map layers. Offline experiments show outstanding reduction in error distance, and online experiments estimated millions in annualized savings.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Before there is any delivery history for an address, the process is bootstrapped by other approximate geocoding methods to guide the driver. Third-party geocodes are not used in our geocode computations.
References
Open Street Map (OSM). www.openstreetmap.org
GIR’10: 6th Workshop on Geographic Information Retrieval. ACM (2010)
ECML/PKDD Competition: Taxi Trajectory Prediction (2015). https://www.kaggle.com/c/pkdd-15-predict-taxi-service-trajectory-i
Aslam, J.A., Kanoulas, E., Pavlu, V., Savev, S., Yilmaz, E.: Document selection methodologies for efficient and effective learning-to-rank. In: SIGIR’09 (2009)
Baraglia, R., Muntean, C.I., Nardini, F.M., Silvestri, F.: Learnext: learning to predict tourists movements. In: CIKM ’13, pp. 751–756 (2013)
de Brébisson, A., Simon, É., Auvolat, A., Vincent, P., Bengio, Y.: Artificial neural networks applied to taxi destination prediction. arXiv:1508.00021 (2015)
Burges, C., et al.: Learning to rank using gradient descent. In: ICML ’05, pp. 89–96 (2005)
Burges, C.J.: From RankNet to LambdaRank to LambdaMart: an overview. Technical report MSR-TR-2010-82, Microsoft Research (2010)
Dang, V.: The Lemur Project-RankLib. https://sourceforge.net/p/lemur/wiki/RankLib/
Hong, L.J., Luo, J., Zhong, Y.: Speeding up pairwise comparisons for large scale ranking and selection. In: IEEE WSC ’16, pp. 749–757 (2016)
Lassoued, Y., Monteil, J., Gu, Y., Russo, G., Shorten, R., Mevissen, M.: A hidden Markov model for route and destination prediction. In: ITSC’17, pp. 1–6 (2017)
Li, H.: SMILE: statistical machine intelligence and learning engine. https://haifengl.github.io/
Lian, D., Xie, X.: Mining check-in history for personalized location naming. ACM Trans. Intell. Syst. Technol. 5(2), 1–25 (2014)
Liu, T.Y.: Learning to Rank for Information Retrieval. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-14267-3
Qin, T., Liu, T.Y., Xu, J., Li, H.: LETOR: a benchmark collection for research on learning to rank for information retrieval. Inf. Retrieval 13(4), 346–374 (2010). https://doi.org/10.1007/s10791-009-9123-y
Rigaux, P., Scholl, M., Voisard, A.: Spatial Databases with Application to GIS. Morgan Kaufmann, Boston (2002)
Ruan, S., et al.: Doing in one go: delivery time inference based on couriers’ trajectories. In: KDD’20, pp. 2813–2821 (2020)
Scott, D.: Multivariate Density Estimation: Theory, Practice, and Visualization. Wiley, Hoboken (1992)
Sculley, D.: Large scale learning to rank. In: NIPS 2009 Workshop on Advances in Ranking (2009)
Shaw, B., Shea, J., Sinha, S., Hogue, A.: Learning to rank for spatiotemporal search. In: WSDM’13, pp. 717–726 (2013)
Wauthier, F., Jordan, M., Jojic, N.: Efficient ranking from pairwise comparisons. In: ICML’13, Atlanta, Georgia, USA, vol. 28, pp. 109–117 (2013)
Ying, J.J.C., Lu, E.H.C., Kuo, W.N., Tseng, V.S.: Urban point-of-interest recommendation by mining user check-in behaviors. In: UrbComp’12, pp. 63–70 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Forman, G. (2021). Getting Your Package to the Right Place: Supervised Machine Learning for Geolocation. In: Dong, Y., Kourtellis, N., Hammer, B., Lozano, J.A. (eds) Machine Learning and Knowledge Discovery in Databases. Applied Data Science Track. ECML PKDD 2021. Lecture Notes in Computer Science(), vol 12978. Springer, Cham. https://doi.org/10.1007/978-3-030-86514-6_25
Download citation
DOI: https://doi.org/10.1007/978-3-030-86514-6_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86513-9
Online ISBN: 978-3-030-86514-6
eBook Packages: Computer ScienceComputer Science (R0)