Show simple item record

dc.contributor.authorZhai, Dongjun
dc.contributor.authorLiu, An
dc.contributor.authorChen, Shicheng
dc.contributor.authorLi, Zhixu
dc.contributor.authorZhang, Xiangliang
dc.date.accessioned2019-05-20T07:10:29Z
dc.date.available2019-05-20T07:10:29Z
dc.date.issued2019-04-24
dc.identifier.citationZhai D, Liu A, Chen S, Li Z, Zhang X (2019) SeqST-ResNet: A Sequential Spatial Temporal ResNet for Task Prediction in Spatial Crowdsourcing. Lecture Notes in Computer Science: 260–275. Available: http://dx.doi.org/10.1007/978-3-030-18576-3_16.
dc.identifier.issn0302-9743
dc.identifier.issn1611-3349
dc.identifier.doi10.1007/978-3-030-18576-3_16
dc.identifier.urihttp://hdl.handle.net/10754/652903
dc.description.abstractTask appearance prediction has great potential to improve task assignment in spatial crowdsourcing platforms. The main challenge of this prediction problem is to model the spatial dependency among neighboring regions and the temporal dependency at different time scales (e.g., hourly, daily, and weekly). A recent model ST-ResNet predicts traffic flow by capturing the spatial and temporal dependencies in historical data. However, the data fragments are concatenated as one tensor fed to the deep neural networks, rather than learning the temporal dependencies in a sequential manner. We propose a novel deep learning model, called SeqST-ResNet, which well captures the temporal dependencies of historical task appearance in sequences at several time scales. We validate the effectiveness of our model via experiments on a real-world dataset. The experimental results show that our SeqST-ResNet model significantly outperforms ST-ResNet when predicting tasks at hourly intervals and also during weekday and weekends, more importantly, in regions with intensive task requests.
dc.description.sponsorshipThis research was supported in part by National Natural Science Foundation of China (NSFC) (Grant No. 61572336, 61632016, 61572335, 61772356), and Natural Science Research Project of Jiangsu Higher Education Institution (No. 18KJA520010, 17KJA520003), and the King Abdullah University of Science and Technology (KAUST), and the Blockshine Technology corp. Data sour Didi Chuxing.
dc.publisherSpringer Nature
dc.relation.urlhttps://link.springer.com/chapter/10.1007%2F978-3-030-18576-3_16
dc.rightsArchived with thanks to Database Systems for Advanced Applications. DASFAA 2019.
dc.subjectDeep neural network
dc.subjectSpatial crowdsourcing
dc.subjectTask prediction
dc.titleSeqST-ResNet: A Sequential Spatial Temporal ResNet for Task Prediction in Spatial Crowdsourcing
dc.typeConference Paper
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.identifier.journalDatabase Systems for Advanced Applications
dc.conference.date2019-04-22 to 2019-04-25
dc.conference.name24th International Conference on Database Systems for Advanced Applications, DASFAA 2019
dc.conference.locationChiang Mai, THA
dc.eprint.versionPost-print
dc.contributor.institutionBlockshine Technology Corp., Shanghai, Blockshine Technology Corp., Shanghai, China, , China
dc.contributor.institutionSoochow University, Suzhou, Soochow University, Suzhou, China, , China
dc.contributor.institutionNational University of Singapore, Singapore, National University of Singapore, Singapore, Singapore, , Singapore
kaust.personZhai, Dongjun
kaust.personZhang, Xiangliang
refterms.dateFOA2020-04-24T00:00:00Z
dc.date.published-online2019-04-24
dc.date.published-print2019


Files in this item

Thumbnail
Name:
378-Liu_Paper.pdf
Size:
1.304Mb
Format:
PDF
Description:
Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record