Log in / Register
Home arrow Computer Science arrow Social Informatics
< Prev   CONTENTS   Next >

4 Data Mining Processes to Trace the Public Life of an Italian Historic City Center

The use and spread of digital social devices has produced a vast amount of data that could be collected in real time maps and give new insights to the way people live within the city.

Nowadays a great optimism is perceived towards the possibilities aroused by the collection and the reading of users' generated data, produced by digital social media activities: these types of observations may require a moment of reflection to understand what significant analyses could be drawn out starting from the interpretation of user's generated data; in particular what analyses could be significant in the case of Urbino and more generally for historic cities centers.

As previously introduced, this research finds particularly relevant to consider data from social media in places like Urbino for the following reasons:

The observable portion of users is significant to the terms of the research. In fact, usually, the main users of digital social media are young people under thirty-five years old2. Thus, students and young native dwellers are fully covered within this age-range.

The results are manageable and thus immediately functional as planning re-

sources. Urbino is a small town, characterized by a condition of spatial isolation. The data mining from digital social media suits particularly well these conditions: no contaminations with exogenous factors are possible and the number of interactions of the inner members is easily controllable and bearable.3

Our main goal was to draw out maps of the social life in public spaces, to incorporate them in a strategy that addresses the whole city center and makes an extensive use of information technologies.

Using data coming from the social network Foursquare we created the Popularity Map and the Attendance Ratio Map, the first registering the most representative places of the city, the latter the type of approach people have towards certain locations (frequently attended venues, occasional venues, special events venues). With the data coming from Flickr, we extracted the Tourist representative poles map, considering what people tend to visit whenever they decide to spend a couple of days in Urbino [11].

Approaching the social network Instagram, the mapping process itself could benefit from the specific physical and social environment of Urbino. In fact Instagram provides several information about the users: apart from the pictures shared – that is a main content itself – it is possible to deduce collateral data concerning the age, gender, profession of the user and where he lives and works.

This enabled us to profile people according to four main categories: the resident, the student – originally not from Urbino, but living and studying in the Marchisan city – the tourist and the commuters – those that even though they are not living inside the historic center of Urbino, reach the city frequently due to working reason or for leisure activities.

The way each Instagram user is assigned to a category follows certain empirical criteria that combine information provided by the users and deducted from the pictures. Applying this methodology of observation to the case of Urbino, by February 2013, it has been possible to identify one hundred sixty Instagram users, which were using the geo-location option in sharing their contents. The profiles have been studied to find out which the most attended places within the historic center were and who the users were. Combining these two inputs we were able to identify which are the most attractive spots of the city for students, tourists, commuters and inhabitants and how their dynamics influence reciprocally.

Fig. 1. Maps of Urbino from social networks data mining. From left: Tourist representative poles Map, Popularity Map, Attendance Ratio Map, Instagram Map (Source: Palmieri, Stojanovic, Tomarchio, Urbino: can digital technology enhance historic public realm?, 2013)

As a result of this assumption, we discovered how certain areas packed with touristic attractions where instead not popular at all among tourists. The other way round, it was possible to map selected places well known among the students, but absolutely not perceived through the sieve of the traditional analysis.

In order to draw these conclusions we crossed data coming from social networks with those that are normally observed in the city with a traditional analysis (eg. mapping building uses, analyzing flows of people and mental maps). Trying to abstract and synthesize the enormous number of potential outcomes of these collected data overlapping, it could be possible to identify three main categories:

Discovery. The digital analysis reveals the existence of certain urban spots that are not detectable by means of traditional analysis.

Matching. The digital analysis spots a series of evidences actually corresponding to

those individuated through the use of the traditional one.

Denial. The digital analysis reports no or little evidences referred to a point that at

a classical observation seems to be a major one. Discovery and Denial areas where those on which the next part of our research focused, exploring what kind of information strategy could be used in a historic city center to enhance its liveability.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science