HA-CMNet: A Driver CTR Model for Vehicle-Cargo Matching in O2O Platform
Paper
Paper/Presentation Title | HA-CMNet: A Driver CTR Model for Vehicle-Cargo Matching |
---|---|
Presentation Type | Paper |
Authors | Jiang, Zilong, Zuo, Xiang, Li, Lin, Tao, Xiaohui and Wang, Dali |
Journal or Proceedings Title | Proceedings of the 19th International Conference on Advanced Data Mining and Applications (ADMA'23) |
Journal Citation | 14179, pp. 648-664 |
Number of Pages | 17 |
Year | 2023 |
Publisher | Springer |
Place of Publication | Switzerland |
ISBN | 9783031466731 |
9783031466748 | |
Digital Object Identifier (DOI) | https://doi.org/10.1007/978-3-031-46674-8_45 |
Web Address (URL) of Paper | https://link.springer.com/chapter/10.1007/978-3-031-46674-8_45 |
Web Address (URL) of Conference Proceedings | https://link.springer.com/book/10.1007/978-3-031-46674-8 |
Conference/Event | 19th International Conference on Advanced Data Mining and Applications (ADMA'23) |
Event Details | 19th International Conference on Advanced Data Mining and Applications (ADMA'23) Parent International Conference on Advanced Data Mining and Applications Delivery In person Event Date 21 to end of 23 Aug 2023 Event Location Shenyang, China Rank B B B B |
Abstract | Vehicle-cargo matching is a key task in freight O2O platform, which involves the complex interactions of drivers, vehicles, cargos, cargo owners and environmental context. Many existing works mainly study the matching of vehicle routing problems, the matching based on the credit evaluation of both drivers and cargo owners, and the matching based on game theory from the perspective of management. Since the freight O2O platform is also the producer of big data, this study proposes a driver CTR prediction model for vehicle-cargo matching task from the perspective of data mining. Specifically, we first use the bottom-level attention network to model fine-grained preferences in driver historical behaviors, such as route interest and search interest, as well as fine-grained preferences such as vehicle type and vehicle length interest, and route interest in cargo owner historical behaviors. Then, the driver basic profile vector, cargo owner basic profile vector, cargo description vector, driver preferences vector and cargo owner preferences vector are feeded into the neural network composed of two deep components for feature interactions learning, and then a top-level attention network is used to learn the influencing factors of different information on the vehicle-cargo matching task. Finally, a multi-classifier is used for matching prediction. We conduct comprehensive experiments on real dataset, and the results show that, compared with the existing solutions, considering user preferences and adopting deep components collaborative modeling can improve the performance of vehicle-cargo matching to a certain extent, which verifies the effectiveness and superiority of the proposed model. |
Keywords | Recommendation system; Online To Offline platform; Vehicle-cargo matching; User preferences modeling; Hierarchical attention network |
Contains Sensitive Content | Does not contain sensitive content |
ANZSRC Field of Research 2020 | 460502. Data mining and knowledge discovery |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
Series | Lecture Notes in Computer Science |
Byline Affiliations | Guizhou University of Finance and Economics, China |
Wuhan University of Technology, China | |
University of Southern Queensland |
https://research.usq.edu.au/item/z9w77/ha-cmnet-a-driver-ctr-model-for-vehicle-cargo-matching-in-o2o-platform
11
total views1
total downloads3
views this month0
downloads this month