ABSTRACT
Estimated Time of Arrival (ETA) plays a vital role in many application scenarios. For example, in various scenarios such as online car-hailing order distribution, price estimation, mid-trip estimation, and route decision-making. Accurate arrival time estimation can help the platform improve efficiency. However, accurate arrival time estimation is affected by static information and dynamic information, and the estimated arrival time has high technical difficulties and challenges.
The organizers of this competition provided departure time and date, itinerary, road conditions, as well as topological structure data and weather information of the city's road network. At the same time, according to the characteristics of the given data, rich feature processing methods such as statistical features, category features, graph features, embedding features, and sequence features are used to provide massive feature information for model learning. One of the most important points is the application of "future data". Of course, in addition to the features, a lot of work has been done on the model structure and model fusion through the combination of machine learning and deep learning, ensuring the accuracy and stability of the model.
The ETA is a typical time series problem. Therefore, In the deep learning section, we choose DCN [3] model and WDR [4] model as the basis, and the model distillation is combined on this as the deep learning part of the integrated model. At the same time, traditional machine learning is also used as a part of the integrated model, through a large number of different dimensions of feature engineering, to make up for the machine learning model's inability to better express the time series problem, and build a machine learning model with higher accuracy. Finally, through the fusion of the deep learning model and the machine learning model, extremely high accuracy is achieved in the ETA problem.
- A.Grover, Jure Leskovec. 2016. Node2vec:Scalable Feature Learning for Networks. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016. https://arxiv.org/abs/1607.00653Google ScholarDigital Library
- G.Hinton, O.Vinyals, J.Dean. 2015. Distilling the Knowledge in a Neural Network. NIPS 2014 Deep Learning Workshop. https://arxiv.org/abs/1503.02531.Google Scholar
- R.Wang, B.Fu, G.Fu, M.wang. 2017. Deep & Cross Network for Ad Click Predictions. In Proceedings of AdKDD and TargetAd, Halifax, NS, Canada, August, 14, 2017, 7 pages. https://arxiv.org/abs/1708.05123.Google ScholarDigital Library
- Z.Wang, K.Fu, J.Ye. 2018. Learning to Estimate the Travel Time. KDD 2018 Poster.Google Scholar
- Elina Parviainen. Deep Bottleneck Classifiers in Supervised Dimension Reduction. Published in ICANN 2010. DOI: https:doi.org/10.1007/978-3-642-15825-4_1.Google Scholar
- Guolin Ke, Qi Meng, Thomas Finley, T.Wang, W.Chen, Weidong Ma, Qiwei Ye, Tie-Yan Liu. LightGBM: A Highly Efficient Gradient Boosting Decision Tree. Advances in Neural Information Processing Systems 30 (NIPS 2017), pp. 3149--3157.Google Scholar
- Takuya Akiba, Shotaro Sano, Toshihiko Yanase, Takeru Ohta, and Masanori Koyama. Optuna: A Next-generation Hyperparameter Optimization Framework. In KDD 2019.Google Scholar
- Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long short-term memory. Neural computation 9, 8 (1997), 1735--1780.Google ScholarDigital Library
- Kyunghyun Cho, Dzmitry Bahdanau, Fethi Bougares Holger Schwenk, Yoshua Bengio. 2014 Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation.Google Scholar
Index Terms
- Integration Model for Estimated Time of Arrival
Recommendations
Retrial queuing system with Markovian arrival flow and phase-type service time distribution
We consider a multi-server queuing system with retrial customers to model a call center. The flow of customers is described by a Markovian arrival process (MAP). The servers are identical and independent of each other. A customer's service time has a ...
Discrete-time single-server finite-buffer queues under discrete Markovian arrival process with vacations
This paper treats a discrete-time single-server finite-buffer exhaustive (single- and multiple-) vacation queueing system with discrete-time Markovian arrival process (D-MAP). The service and vacation times are generally distributed random variables and ...
Performance prediction of a discrete-time batch arrival retrial queue with Bernoulli feedback
This paper examines a discrete-time GeoX/G/1 retrial queue wherein the customer is feedback again to the head of the queue with some probability in case when he is unsatisfied with his service. This phenomenon is called as Bernoulli feedback. We ...
Comments