Urban Computing:Enabling Intelligent Cities with AI and Big Data

Dr./Prof. Yu Zheng 
Senior Research Manager, Microsoft Research, Beijing China
Editor-in-Chief of ACM Transactions on Intelligent Systems and Technology
Email: yuzheng@microsoft.com Phone: +86-10-59173038
Address: Suit 301, World Trade and Convention Centre, Halifax, Canada
Homepage of Urban Computing
(A Half-Day Tutorial)
Download the slide deck
Short bio of the speaker
Dr. Yu Zheng is a senior research manager in Urban Computing Group at Microsoft
Research. His publications have been cited over 14,000 times (Google Scholar H-Index 55).
He currently serves as the Editor-in-Chief of ACM Transactions on Intelligent Systems and
Technology, and on the editorial board of IEEE Transactions on Big Data. He is also a
founding secretary of SIGKDD China Chapter. Zheng has chaired over 10 prestigious
international conferences, such as the program co-chair of ICDE 2014 (Industrial Track) and
Industrial PC co-chair of CIKM 2017, and served as senior PCs at KDD, ICDM, and SDM. He is the founder
of the SIGKDD Workshop on Urban Computing and a key organizer of the past five rounds of this
workshop. He has been a keynote speaker over 10 IEEE/ACM international conferences. In 2013, he
was named one of the Top Innovators under 35 by MIT Technology Review (TR35) and featured by
Time Magazine for his research on urban computing. In 2016, Zheng was named ACM Distinguished
Scientist for his contribution to spatio-temporal data mining and urban computing. Zheng is also a Chair
Professor at Shanghai Jiao Tong University, an Adjunct Professor at Hong Kong University of Science
and Technology, and Hong Kong Polytechnic University.
Abstract
Urban computing is a process of acquisition, integration, and analysis of big and heterogeneous
data generated by a diversity of sources in cities to tackle urban challenges, e.g. air pollution,
energy consumption and traffic congestion. Urban computing connects unobtrusive and ubiquitous
sensing technologies, advanced data management and analytics models, and novel visualization
methods, to create win-win-win solutions that improve urban environment, human life quality,
and city operation systems. Urban computing is an inter-disciplinary field where computer science
meets urban planning, transportation, economy, the environment, sociology, and energy, etc.,
in the context of urban spaces [1][2]. The vison of urban computing has been leading to better cities
that matter to billions of people.
Though the concept of urban computing has been proposed for a few years, there are still quite a few
questions open. For example, what are the core research problems of urban computing? What are the
challenges of the research theme? What are the key methodologies for urban computing? What are the
representative applications in this domain, and how does an urban computing system work?
Outline
In this tutorial, I will overview the framework of urban computing, discussing its key challenges and
methodologies from data science’s perspective (particularly from data mining’s perspective). This
tutorial will also present a diversity of urban computing applications, ranging from big data-driven
environmental protection [15][12] to transportation[13], from urban planning to urban economy [14].
The research has not only published at prestigious conferences like KDD, but also deployed in the real
world. Here is the outline:
● Urban sensing
  ○ Challenges and key techniques
  ○ Filling missing values in geo-sensory data [4]
  ○ Resource allocation in urban sensing [3][6]
● Urban data management
  ○ Urban big data platform: a framework and key components [8]
  ○ Trajectory data management on the cloud [5]
● Urban data analytics
  ○ Key challenges and techniques
    ◾ Machine learning for spatio-temporal data
    ◾ Combining data management with machine learning organically
    ◾ Cross-domain knowledge fusion
    ◾ Visual and interactive data analytics
  ○ Cross-domain knowledge fusion [9]
    ◾ Stage-based methods
    ◾ Feature-based knowledge fusion
      • Feature concatenation + regularization [14]: One examples is ranking
        urban real estates based on a diversity of data
      • Deep learning-based fusion methods [10]: One example is to forecast the
        flow of crowds based on deep learning and multiple data sources
    ◾ Semantic meaning-based knowledge fusion
      • Multi-view-based methods [15][12]: One example is to predict fine-
        grained air quality throughout a city based on Multiview learning
      • Probabilistic dependency-based methods [16][18]: One example is to infer
        the gas consumption and pollution emission of vehicles on each and every
        road in a city based on graphical models.
      • Similarity-based methods [13][17]: One example is to estimate the travel
        time of a path based on sparse trajectories.
      • Transfer learning-based fusion methods [11]: One example is to transfer
        knowledge from cities with sufficient data to others without enough data.
Target audience of this tutorial
Graduate students, professors and professionals working on the research and practices on smart cities
and spatio-temporal data mining, with very basic data mining concepts, such as frequent pattern mining,
regression, classification and clustering. Note that there are already quite a few universities, e.g. NYU,
which have setup master degrees on urban computing or urban informatics. I have signed contract with
MIT Press to prepare the first text book for this area. The book will be published later 2017. The main
content of this book comes from this tutorial. Thus, the tutorial will be well organized and easy to
follow.
History of the tutorial
This tutorial has not been presented at any conferences so far, but will be presented at DASFAA 2017
on March 27, 2017 as per the PC chair’s invitation. The tutorial I proposed to KDD 2017 will share the
most content with that at DASFAA 2017, with some new research results added. A part of the content
of this tutorial has been presented in my keynote speeches at the following IEEE/ACM conferences and
workshops (thus, there is a guarantee for its quality):
● Keynote Speech at the 7th ACM SIGSPATIAL International Workshop on GeoStreaming (IWGS
  2016), San Francisco, USA, 2016.10.31.
● Keynote Speech at the 24th IEEE International Requirements Engineering Conference (RE 2016),
  Beijing China, 2016.9.15
● Keynote Speech at the 10the IEEE International Conference on Big Data Science and Engineering
  (IEEE BIGDataSE 2016), Tianjin, China, 2016.8.24
● Keynote Speech at the 5th ACM SIGKDD International Workshop on Urban Computing
  (UrbComp 2016), San Francisco, USA, 2016.8.14
● Keynote Speech at the 4th ACM SIGKDD International Workshop on Urban Computing
  (UrbComp 2015), Sydney, Australia, 2015.8.10
● Keynote Speech at the 2nd International Conference on Sustainable Urbanization (ICSU 2015),
  Hong Kong, 2015.1.
● Keynote speech at the 10th IEEE International conference on Intelligent Environment (IE 2014),
  Shanghai, China, July 2014.
References
Publications related to the framework of urban computing:
[1] Yu Zheng, Licia Capra, Ouri Wolfson, Hai Yang,Urban Computing: Concepts, Methodologies, and
  Applications, in ACM Transaction on Intelligent Systems and Technology
[2] Yu Zheng, Urban Computing. MIT Press, a text book to appear in later 2017
Publications related to urban sensing:
[4] Xiuwen Yi, Yu Zheng, Junbo Zhang, Tianrui Li, ST-MVL: Filling Missing Values in Geo-sensory Time
  Series Data, IJCAI 2016
Publications related to urban data management:
[5] Jie Bao, Ruiyuan Li, Xiuwen Yi, Yu Zheng. Managing Massive Trajectories on the Cloud . ACM
  SIGSPATIAL 2016
[6] Yuhong Li, Jie Bao, Yanhua Li, Yingcai Wu, Zhiguo Gong, Yu Zheng. Mining the Most Influential k-
  Location Set from Massive Trajectories. ACM SIGSPATIAL 2016
  Locations”, IEEE Transactions on Visualization and Computer Graphics
[8] Yu Zheng, Trajectory Data Mining: An Overview , in ACM Transactions on Intelligent Systems and
  Technology (ACM TIST)
Publications related to urban data analytics:
[9] Yu Zheng, Methodologies for Cross-Domain Data Fusion: An Overview , IEEE Transactions on Big Data
  Flows Prediction, AAAI 2017.
[11] Ying Wei, Yu Zheng, Qiang Yang, Transfer knowledge between Cities, KDD 2016
[12] Yu Zheng, Xiuwen Yi, Ming Li, Ruiyuan Li, Zhangqing Shan, Eric Chang, Tianrui Li, Forecasting Fine-
[13] Jingbo Shang, Yu Zheng, Wenzhu Tong, Eric Chang, Yong Yu, Inferring Gas Consumption and
[14] Yanjie Fu, Hui Xiong, Yong Ge, Zijun Yao, Yu Zheng, Zhi-Hua Zhou, Exploiting Geographic
[15] Yu Zheng, Furui Liu, Hsun-Ping Hsieh, U-Air: When Urban Air Quality Inference Meets Big Data
  , KDD 2013
[16] Hsun-Ping Hsieh, Shou-De Lin, Yu Zheng, Inferring Air Quality for Station Location Recommendation
  Based on Urban Big Data, in Proceedings of the 21th SIGKDD conference on Knowledge Discovery and
  Data Mining (KDD 2015)
[17] Yilun Wang, Yu Zheng, Yexiang Xue, Travel Time Estimation of a Path using Sparse Trajectories, in
  Proceedings of the 20th SIGKDD conference on Knowledge Discovery and Data Mining (KDD 2014)
[18] Yu Zheng, Huichu Zhang, Yong Yu, Detecting Collective Anomalies from Multiple Spatio-Temporal
  Datasets across Different Domains, in Proceedings of the 23rd ACM International Conference on
  Advances in Geographical Information Systems (ACM SIGSPATIAL 2015)