A Neighborhood-Augmented LSTM Model for Taxi-Passenger Demand Prediction

verfasst von

Tai Le Quy, Wolfgang Nejdl, Myra Spiliopoulou, Eirini Ntoutsi

Abstract

Taxi is a convenient means of transportation worldwide. Accurately predicting the taxi-demand is crucial for taxi-companies to effectively allocate their fleet to taxi-stands and reduce the waiting time for passengers thus increasing their overall satisfaction and customer retention. Nowadays precise information about taxi-rides is available and can be used to infer the taxi-passenger demand across different locations and time-points. In this paper, we propose an approach for predicting the pick-demand of a given taxi-stand, that takes into account not only the demand-history of the particular stand but it also considers information from neighboring stands. Our model is an LSTM neural network augmented with information from the spatial neighborhood of the stands. Experiments with two versions of the taxi demand dataset from the city of Porto, Portugal show that our approach can provide better predictions comparing to approaches that do not exploit the neighborhood.

Organisationseinheit(en)

Forschungszentrum L3S

Externe Organisation(en)

Otto-von-Guericke-Universität Magdeburg

Typ

Aufsatz in Konferenzband

Seiten

100-116

Anzahl der Seiten

17

Publikationsdatum

04.01.2020

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Theoretische Informatik, Informatik (insg.)

Ziele für nachhaltige Entwicklung

SDG 11 – Nachhaltige Städte und Gemeinschaften

Elektronische Version(en)

https://doi.org/10.1007/978-3-030-38081-6_8 (Zugang: Offen)