报告题目： Torch: a Search Engine for Trajectory Data
报告人单位：皇家墨尔本理工 （RMIT University）, 澳大利亚
Zhifeng Bao is an Associate Professor in Computer Science, RMIT (Royal Melbourne Institute of Technology) university and an Adjunct Fellow at University of Melbourne, Australia. He received his PhD from the CS Dept at NUS in 2011. Zhifeng was the only recipient of the Best PhD Thesis Award in School of Computing and was the winner of the Singapore IDA (Infocomm Development Authority) gold medal. Zhifeng was a winner of the Google Faculty Research Award 2015. His research interests include data visualization, spatial data analytics for smart transportation and tourism, machine learning and graph data management. He served the PC Co-chair of DASFAA17, ER18, APWEB16, WSDM19 Cup, etc, and served the PC member of top conferences such as VLDB17-18, SIGMOD18, SIGIR15-18, ICDE16-19, IJCAI16. Zhifeng has received four best paper awards such as DASFAA17, ADC16, and five best paper nomination such as IEEE ICDE 2009, CIKM 2014. Since 2015 he has secured more than 1 million AUD funding as the chief investigator from Australasian Research Council, CSIRO and Google.
In this talk I will introduce our recent work accepted by ACM SIGIR 2018. This paper presents a new trajectory search engine called Torch for querying road network trajectory data. Torch is able to efficiently process two types of typical queries (similarity search and Boolean search), and support a wide variety of trajectory similarity functions. Additionally, we propose a new similarity function LORS in Torch to measure the similarity in a more effective and efficient manner. Indexing and search in Torch works as follows. First, each raw vehicle trajectory is transformed to a set of road segments (edges) and a set of crossings (vertices) on the road network. Then a lightweight edge and vertex index called LEVI is built. Given a query, a filtering framework over LEVI is used to dynamically prune the trajectory search space based on the similarity measure imposed. Finally, the result set (ranked or Boolean) is returned. Extensive experiments on real trajectory datasets verify the effectiveness and efficiency of Torch.