DAMASCA: DAta Mining And Smart Cities Applications Workshop 2015

DAMASCA Workshop Program

Time and Place: Mon. Apr. 13 at 307B, COEX

9:00 - 9:10 Introduction (Veli Bicer)

9:10 - 9:50 Keynote Speech 1 : Smart City and Big Data : with the case of Korea (Dr. Jong-Sung Hwang)

• This talk is going to deal with three issues on achieving a smart city vision. The first issue is about the definition in which I will define smart city as "City as a Platform" enabling the exchange and analysis of a variety of data on city from many different sources like sensors, administration, and people. The second is the strategy, or how to realise smart city. We can identify two approaches; one emphasises the IoT side and the other the data analytics side. I will explain strength and weakness of each based on Korean experiences. The third issues is the value of smart city, or how to maximise the value of smart city services. I'll discuss some cases from Seoul as well as other cities.

9:50 - 10:20 Smart City and Geospatiality: Hobart Deeply Learned (Jagannath Aryal, Ritaban Dutta)

10:20 - 11:00 Coffee Break

11:00 - 11:30 Managing travels with PETRA: the Rome use case (Adi Botea, Stefano Braghin, Nuno Lopes, Riccardo Guidotti, Francesca Pratesi)

11:30 - 12:00 Monitoring the citizens' perception on urban security in Smart City environments (Luca Cagliero, Silvia Chiusano, Tania Cerquitelli, Luca Venturini, Marco Nardone, Barbara Pralio, Pierangelo Garino)

12:00 - 12:30 POP: a Passenger-Oriented Partners Matching System (Xiaoyi Duan, Cheqing Jin, Xiaoling Wang)

12:30 - 14:00 Lunch

14:00 - 14:40 Keynote Speech 2 : Statistical approaches for Non-Intrusive Load Monitoring over IoT platform (Dr. Ji Won Yoon)

• Non-Intrusive Load Monitoring (NILM) is a key process to build desired smart cities over IoT platform in that it identifies what appliances are running in a house by analyzing the power consumption signals or related side channels of a smart meter. In this talk, we first show why NILM is a difficult process in a statistical point of view. After presenting several approaches including our approach for NILM, several potential applications using NILM techniques will be discussed. For example, NILM system can effectively be used for designing smart healthcare, energy-saving, and security.

14:40 - 15:10 Data Mining for Safety Transportation by Means of Using Internet Survey (Masahiro Miyaji)

15:10 - 15:40 On Crowdsensed Data Acquisition using Multi-Dimensional Point Processes (Saket Sathe, Timos Sellis, Karl Aberer)

15:40 - 16:00 Coffee Break

16:00 - 16:30 Discovering Underpasses from Walking Trajectories (Qiuge Song, Cheqing Jin, Xiaoling Wang, Ming Gao, Aoying Zhou)

16:30 - 17:00 Hotel Recommendation based on User Preference Analysis (Kai Zhang, Keqiang Wang, Xiaoling Wang, Cheqing Jin)

17:00 - 17:30 Discussion & Closing

About

In today’s urbanizing world, cities have not only physical infrastructures, such as road networks, utilities, or buildings, but also comprise a knowledge infrastructure ranging from lowlevel sensor networks to public databases and social media streams. The data emerging from all those sources is a very precious resource to make cities more intelligent, innovative and integrated beyond the boundaries of isolated applications. Although such “big data” has been popularised in the media as the “new currency", fuelling a future vision of contextual systems that will transform our cities, the reality is that we just began to recognise significant research challenges across a spectrum of topics (that must be addressed to realise the vision: information retrieval, knowledge representation, semantic reasoning, data mining and many others). In fact, if we cannot find the ways to harvest the city data and to transform it into tangible insights, our vision of offering innovative solutions to the realworld problems of the cities will not go beyond an expressed wish.

This workshop addresses such a timely issue to turn the city data into insightful information. It covers a broad range of topics rooting from different scientific fields, in order to enable novel research to mine important patterns from city data and to apply them in various emerging application areas, such as smart mobility/transportation, smart tourism or smart participation. Those application areas pose unique challenges giving the opportunity to researchers in different communities, including database, knowledge management and information retrieval, to discuss the emerging research topic of pattern mining in smart cities, identifying its unique challenges, opportunities and future directions.

Objectives

The workshop aims at providing a forum to discuss challenges and opportunities of pattern mining in smart cities and to promote interaction among experts in that area. In this regard, one of the main objectives includes the positioning of the importance of pattern mining in the context of cities among the other types of data mining and analytics, by identifying the unique characteristics and requirements of city data (e.g. heterogeneity, openness, geospatiality, privacy issues, etc.). Another objective is also to show how the particular challenges of cities (e.g. urban planning, resource optimization, emergency evacuation, transportation planning) can be better addressed by exploiting the knowledge extracted from the vast amount of city data. Then, drawing a parallel between the topics of this workshop and the topics of the major tracks of ICDE will be a third objective of the workshop. The main outcome will be to produce a roadmap and a community of researchers interested in this topic in order to guide further developments and technological progress towards smarter cities.

Topics of Interest

In order to better address the challenges in enabling insightful information in a heterogeneous city environment, we welcome topics grouped in three categories: The first category addresses the specific techniques in different areas of data mining and analytics that enable the discovery of patterns from heterogeneous city data. In the second category, we will focus on innovative techniques for using discovered patterns to enable missioncritical information to various city stakeholders via a seamless information access. Finally, the workshop aims to discuss the usage of the various techniques in particular application scenarios with the aim of revealing specific challenges of each application area. This workshop targets (but is not limited to) the following topics, grouped into these three categories:

Data mining and analytics

• Data mining for smart cities
• Mining citywide streams
• Geospatial analysis of city data
• Text mining, opinion mining and sentiment analysis on city data
• Social network analysis
• Clustering, classification, and summarization of city data
• Predictive analysis for optimization of city infrastructure and city resiliency
• Mobile data mining
• Warehousing heterogeneous city data
• Environmental data mining

Information Access

• Complex Event Processing for Smarter Cities
• Anomaly detection and prevention
• Forecasting city events
• Eventbased optimization for adaptive city operations
• City process monitoring
• Cloud Computing for pattern mining in smart cities
• Modeling and simulation of city infrastructure

Application Scenarios

• Smart mobility and transportation
• Smart tourism
• Smart participation
• Smart environment
• Smart energy
• Smart water
• Smart Locationbased services
• Smart emergency management

Organizers

• Michele Berlingerio, IBM Research - Smarter Cities Technology Centre, Dublin, Ireland
• Veli Bicer, IBM Research - Smarter Cities Technology Centre, Dublin, Ireland
• Mirco Nanni, KDDLab CNR Pisa, Italy
• Roberto Trasarti, KDDLab CNR Pisa, Italy

Preliminary Program Committee

• Gennady Andrienko (UCL)
• Mirco Musolesi (University of Birmingham)
• Adi Botea (IBM)
• Andrea Passarella (IITCNR)
• Dino Pedreschi (University of Pisa)
• Raffaele Perego (ISTICNR)
• Fabio Pinelli (IBM Research Dublin)
• Salvatore Rinzivillo (ISTICNR)
• Zbigniew Smoreda (Orange)
• Philip Yu (UIC)

Submission and Proceedings

For submissions, the following rules apply:

• Full technical papers: up to 5 pages in IEEE format
• Short position or demo papers: up to 2 pages in IEEE format

Submissions must be formatted using the IEEE templates available here. Submissions will be peer reviewed by three independent reviewers. Accepted papers will be presented at the workshop and published jointly with the conference proceedings. We will pursue a journal special issue with the topics of the workshop if we receive an appropriate number of high-quality submissions. Details on the proceedings and camera-ready formatting will be announced upon notification of the authors. Please use the following link to the submission system to submit your paper at Easychair Submission System for DAMASCA

Contact

The organization committee can be reached using the contact information above.