Secure Your Ride: Real-time Matching Success Rate Prediction for Passenger-Driver Pairs

نویسندگان

چکیده

In recent years, online ride-hailing platforms, such as Uber and Didi, have become an indispensable part of urban transportation make our lives more convenient. After a passenger is matched up with driver by the platform, both freedom to simply accept or cancel ride one click. Hence, accurately predicting whether passenger-driver pair good match, i.e., its matching success rate (MSR), turns out be crucial for platforms devise instant strategies order assignment. However, since users consist two parties, decision-making needs simultaneously account dynamics from sides. This makes it challenging than traditional advertising tasks that predict user's response towards object, e.g., click-through prediction advertisements. Moreover, amount available data severely imbalanced across different cities, creating difficulties training accurate model smaller cities scarce data. Though sophisticated neural network architecture can help improve accuracy under scarcity, overly complex design will impede model's capacity delivering timely predictions in production environment. paper, MSR passenger-driver, we propose M ulti- xmlns:xlink="http://www.w3.org/1999/xlink">V iew ( MV ) which comprehensively learns interactions among dynamic features passenger, driver, trip order, well context. Regarding imbalance problem, further xmlns:xlink="http://www.w3.org/1999/xlink">K nowledge xmlns:xlink="http://www.w3.org/1999/xlink">D istillation framework xmlns:xlink="http://www.w3.org/1999/xlink">KD supplement predictive power using knowledge denser data, also generate simple support efficient deployment. Finally, conduct extensive experiments on real-world datasets several demonstrates superiority solution.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

RoadMusic : Music For Your Ride From Your Ride

RoadMusic is an artistic project, which generates music for in-car listening, from information gathered while driving. We will describe in detail how the system works, and briefly define our intentions in initiating this project. We will demonstrate how, through the musical format that is generated from the data, there is a possible subliminal perception of the situation. We will suggest ways i...

متن کامل

State-feedback Control for Passenger Ride Dynamics

An analytical investigation of a half-car model with passenger dynamics, subjected to random road disturbance, is performed. Two diierent methods of deening the performance index for optimal controller design are proposed. Nondeterministic inputs are applied to simulate the road surface conditions more realistically. Results obtained illustrate that using an optimal state-feedback controller, w...

متن کامل

Dancing with your ride

The technologies that make this possible are those of miniaturized electronics, digital communications, inexpensive distributed computation, and advanced control software. They enable us to get rid of most of the junk that currently encumbers automobiles – engines, power trains, dashboards, and steering wheels. They allow us to define a fundamentally new, radically restructured architecture for...

متن کامل

Passenger Ride Comfort through Observer Based Control

abstract An analytical investigation of a half-car model including passenger dynamics, subjected to random road disturbances is performed, and the advantage of active over conventional passive suspension systems are examined. Two diierent performance indices for optimal controller design are proposed. The performance index is a quantiication of both ride comfort and road handling. Due to practi...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE Transactions on Knowledge and Data Engineering

سال: 2022

ISSN: ['1558-2191', '1041-4347', '2326-3865']

DOI: https://doi.org/10.1109/tkde.2021.3112739