Mining spatial data has been a core subject of study in the data mining community over the past years. Most of scholarly research has focused on the analysis of GPS traces and place recommendations. More recently however, new layers (e.g. social, semantic, linguistic) of big location data have emerged. Given the unprecedented levels of urbanization experienced in the last decade, among the most challenging and crucial ones is the urban fabric layer. The latter includes information that ranges from data related to transportation and navigation in a city to data that are related with the local economy. To integrate urban studies with the research agendas revolving around traditional data mining conferences, it has become clear that a basic introduction to urban studies is needed. The goal of this tutorial is twofold; (a) to provide this introduction in a form that is focused on topics most relevant to the ICWSM community and, (b) to introduce its attendants to the state-of-the-art in the analysis and modeling in this new regime of spatial data with a special focus on urban applications.
In the first part of the tutorial, Daniele, covers urban and social theories developed in the past 50 years including the gravity model, phsycological maps and the notion of aesthetic capital. He further presents an application of each theory and ways that we can utilize Web technologies to validate these theories and further exploit them for applications such as navigation algorithms target happy pedestrian routes.
Presentation slides: (pdf)
In the second part of the tutorial, Kostas covers basic techniques useful for analyzing the highly heterogenous urban data that originate from a variety of data sources. The focus is on matrix and tensor factorization techniques that can provide infromation for latent urban activity patterns. He will discuss and cover the basic matrix and tensor factorization techniques and provide pointers to existing literature in urban informatics that makes use of them.
Presentation slides: (pdf)
In this part of the tutorial, Tassos, will present data mining and statistical modeling techniques in the context of geographic data analysis and human mobility modeling. Furthermore, urban activity modeling will be covered with a specific focus on urban density and temporal dynamics. We will also discuss emerging research on topics related with local businesses and urban economy pertaining to optimal retail store placement and advertisement through social media, while we will finally introduce new analysis, modeling and application opportunities rising in the changing field of urban taxi transport.
Presentation slides: (pdf)
In this session Bruno will start by very briefly introducing the Twitter platform and detailing the demographics of the users and the biases they introduce. The relationship between geography, mobility and social network properties will be described using the Twitter service as a case study. Next, the classic link prediction problem will be discussed in the context of geographic and location-based social networks. Finally, information spreading is an important process in online social networks that has attracted the interest of many researchers over the past decade. In this part of the tutorial attendees will get the chance to review the most seminal works in the area where spatial and geographic perspectives are highlighted.
Presentation slides: (pdf)