Show simple item record

dc.contributor.author
Tong, Yongxin
dc.contributor.author
Chen, Yuqiang
dc.contributor.author
Zhou, Zimu
dc.contributor.author
Chen, Lei
dc.contributor.author
Wang, Jie
dc.contributor.author
Yang, Qiang
dc.contributor.author
Ye, Jieping
dc.contributor.author
Lv, Weifeng
dc.date.accessioned
2017-12-13T10:33:52Z
dc.date.available
2017-10-06T05:02:12Z
dc.date.available
2017-12-13T10:33:52Z
dc.date.available
2017-12-05T08:08:46Z
dc.date.issued
2017-08
dc.identifier.isbn
978-1-4503-4887-4
en_US
dc.identifier.other
10.1145/3097983.3098018
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/220494
dc.description.abstract
Taxi-calling apps are gaining increasing popularity for their eficiency in dispatching idle taxis to passengers in need. To precisely balance the supply and the demand of taxis, online taxicab platforms need to predict the Unit Original Taxi Demand (UOTD), which refers to the number of taxi-calling requirements submitted per unit time (e.g., every hour) and per unit region (e.g., each POI). Predicting UOTD is non-trivial for large-scale industrial online taxicab platforms because both accuracy and fexibility are essential. Complex non-linear models such as GBRT and deep learning are generally accurate, yet require labor-intensive model redesign after scenario changes (e.g., extra constraints due to new regulations). To accurately predict UOTD while remaining fexible to scenario changes, we propose Lin UOTD, a unifed linear regression model with more than 200 million dimensions of features. The simple model structure eliminates the need of repeated model redesign, while the high-dimensional features contribute to accurate UOTD prediction. We further design a series of optimization techniques for eficient model training and updating. Evaluations on two largescale datasets from an industrial online taxicab platform verify that Lin UOTD outperforms popular non-linear models in accuracy. We envision our experiences to adopt simple linear models with high-dimensional features in UOTD prediction as a pilot study and can shed insights upon other industrial large-scale spatio-temporal prediction problems. © 2017 ACM.
en_US
dc.language.iso
en
en_US
dc.publisher
ACM
en_US
dc.subject
Feature engineering
en_US
dc.subject
Prediction
en_US
dc.subject
Unit original taxi demands
en_US
dc.title
The Simpler The Better: A Unified Approach to Predicting Original Taxi Demands based on Large-Scale Online Platforms
en_US
dc.type
Conference Paper
ethz.book.title
Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '17
en_US
ethz.journal.volume
Part F129685
en_US
ethz.pages.start
1653
en_US
ethz.pages.end
1662
en_US
ethz.event
23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2017)
en_US
ethz.event.location
Halifax, NS, Canada
en_US
ethz.event.date
August 13-17, 2017
en_US
ethz.identifier.scopus
ethz.publication.place
New York, NY, USA
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02640 - Inst. f. Technische Informatik und Komm. / Computer Eng. and Networks Lab.::03429 - Thiele, Lothar / Thiele, Lothar
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02640 - Inst. f. Technische Informatik und Komm. / Computer Eng. and Networks Lab.::03429 - Thiele, Lothar / Thiele, Lothar
en_US
ethz.date.deposited
2017-10-06T05:02:13Z
ethz.source
SCOPUS
ethz.source
FORM
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2017-12-13T10:36:43Z
ethz.rosetta.lastUpdated
2018-11-06T05:30:26Z
ethz.rosetta.versionExported
true
dc.identifier.olduri
http://hdl.handle.net/20.500.11850/217187
dc.identifier.olduri
http://hdl.handle.net/20.500.11850/192407
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=The%20Simpler%20The%20Better:%20A%20Unified%20Approach%20to%20Predicting%20Original%20Taxi%20Demands%20based%20on%20Large-Scale%20Online%20Platforms&rft.date=2017-08&rft.volume=Part%20F129685&rft.spage=1653&rft.epage=1662&rft.au=Tong,%20Yongxin&Chen,%20Yuqiang&Zhou,%20Zimu&Chen,%20Lei&Wang,%20Jie&rft.isbn=978-1-4503-4887-4&rft.genre=proceeding&rft_id=info:doi/978-1-4503-4887-4&rft.btitle=Proceedings%20of%20the%20ACM%20SIGKDD%20International%20Conference%20on%20Knowledge%20Discovery%20and%20Data%20Mining,%20KDD%20'17
 Search via SFX

Files in this item

FilesSizeFormatOpen in viewer

There are no files associated with this item.

Publication type

Show simple item record