Challenging Methods and Results Obtained from User-Generated Content in Barcelona’s Public Open Spaces

The implications of User Generated Content are changing the daily life of people. Everyone is involved, but some institutions have stored the data and the power to exploit them. We can contribute to make this power more democratic and available to improve the life of people. This chapter has the goal to do it. The objective of this chapter is to deepen the study on user-generated data (UGD, User Generated Content, UGC) for exploring new methodologies that could support and improve the understanding of spatial patterns for urban planning and design of open spaces in urban areas. This chapter aims at taking a step forward form current literature. It provides, on the one hand, three methods for social science analysis using Big Data; and, on the other hand, it expects to motivate the possible outcomes of UGD coming from three Information and Communication Technologies (ICT) platforms into policy driving strategies, where results can be used to improve quality of life. This chapter examines the spatial pattern of UGD in three public open spaces in Barcelona: Enric Granados Street (Carrer Enric Granados), Forum (Fòrum de les Cultures) and Barcelona Metropolitan Area; and the spatial patterns generated by weekday and by hourly interval in order to study the further insides of how users construct different configuration in visiting a city. Methodologically, we have used two different platforms (Twitter and WAY app). Twitter, the web social network is currently used by 284 million of monthly active users (Twitter Inc. 2015). Twitter provides large amounts of data georeferenced. Although only a part of the entire urban population uses Twitter, the remaining data are a useful source to analyse the location of people in the city are and why there is more activity-clustering in places vs. others. The second platform is the app WAY Cyberparks, an application developed by DeustoTech Mobility within the COST Action TU1306 – CyberParks (Bahillo et al. 2016, 2015; Masegosa et al. 2015). This GPS-based platform focuses on the use of ICT for understanding and collecting data on the use of urban open spaces.

The possibility to communicate worldwide through common smartphones opened up new perspectives and interaction pathways to individuals, who very quickly responded with a massive use of social network platforms and applications. Since most smartphones include Global Positioning System (GPS), if activated, outgoing data from each device can be geo-referenced. Thus, the large amount of data can be geo-located with a time reference. The consequence is the constant production of “Big Data” (Targio et al. 2015; Snijders et al. 2012) which can be overlaid on maps and analysed by mathematical, semantic techniques (Language detection API 2016), and by means of spatial analysis methods. Substantial works on the nature of geography have discussed the spatial pattern formed by many purposes of analysis, since the classical works of Hartshorne (1939), Harvey (1963), Friedman and Alonso (1964), Berry (1973), and Chorley and Haggett (1965) among many others till nowadays. Their relevant research on the organization and settlement of people and activities in the human world, the regularity of distribution brought significant and worthy results which made huge advancements in Geographical Thought and society. In 2018, UGC brought a new paradigm in the study and analysis of spatial patterns in Human Geography. Related problems are more associated to find appropriate models and methodologies to draw a good statistical sample from the population, to depurate the data, to discriminate the data size and to analyse consistently the results.

Case studies are very specific and different POS in Barcelona. Enric Granados is located at the urban centre, in the “Cerdà’s Eixample” (Cerdà enlargement) of Barcelona (Pallares-Barbera, Badia and Duch 2011). It is a residential lively, dense street with services and amenities; Forum is located at the end of the Diagonal, very close to the 22@Barcelona neighbourhood (Casellas and Pallares-Barbera 2009). It is mainly a complex of buildings and cement plazas built for the 2004 Forum of Cultures exhibition. For the most part, locals use Enric Granados Street, while Forum’s facilities are used for large meetings, conventions and festivals; otherwise it is an unoccupied space.

This chapter is formulated in different sections. Section 2 analyses the state of the art regarding uses and applications of UGC by different institutions, firms and policy makers. Section 3 describes the case studies. Section 4 focusses on data and methodology, while Sect. 5 is about results and discussion of them. Conclusions, provided in Sect. 6, aim to offer insides of expected future developments for the integration of UGC with new methodologies within contemporary spatial planning processes.

The availability of big and open data is nowadays offering new opportunities for re-thinking the human behaviour and its impact on Earth. Since these data is often geo-referenced, its overlaying on maps provides useful information on particular organization and structure of spatial systems. Therefore, urban planning, and design, transport planning, social sciences and human geography can benefit from their use.

Although a large number of applications and experiments constellates the landscape of research, the academic research has barely yet found a specific own pathway within these huge opportunities. However, relevant academic work is coming out as well as public and private bodies have begun to use big data for their own development; for example, NOAA Big Data Project, or DataKind for the treatment of trees in NYC. A concise overview on the state of the art on the use of geolocated user-generated contents in spatial planning is meant to show analysis, applications and uses.

Cities have the pressure of millions of inhabitants to increase their quality of life. Through the introduction of ICT, cities try to cope population needs with strategies based on an integrative framework, including management, technology, governance and policy; a reunion of factors which some call smart city (Chourabi et al. 2012). The concept of smart city has not a clear consensus and nor a consistent understanding among academia and practitioners, but in this chapter, it will be used to refer to a city with massive implications of ICT. City information is collected by sensors, tools and applications and used on the city policy. Data-information relies on the diffuse use of devices for quantifying movements, environment quality and healthiness, and energy consumption. Private enterprises, such as the biggest technological companies like IBM, Cisco or Siemens are also interested on data collection and analysis for their own objectives and strategies.

Local institutions such as governments or urban planning firms use big data to organize exhibitions with different goals. Some of them are under the idea of crowd-sourcing, in which citizens’ digital footprint such as tweets or use of online municipal services help city planners improve city’s design and functionality. For instance, the Chicago Architecture Foundation (CAF) partnering with IBM (Chicago Architecture Foundation 2014; IBM 2014; Snodgrass 2014) launched “Chicago: City of Big Data.” Its goal was to open up a dialogue on the connection between big data and human lives and to show how spatial systems are impacted by this information, and how relevant each individual was to the design of their environment. The exhibition’s dashboards were built with IBM’s City Forward platform, a free web-based “civic resource that enables people to visualize and interact with city data” (Kokalitcheva 2014). The methodological basis for selecting data sources used two central criteria: the data’s availability at a large enough scale; and its capability to communicate valuable insights about the city’s needs and flows. Then, they used complex geographic information systems (GIS) and web-based tools to compile and visualize the data in effective ways. Data sources consisted of over 18 million Twitter data points, with only timestamp and location attributes, which turned into dynamic and “interactive models of city’s human activity.” (Kokalitcheva 2014). On the broad objective of “explor(ing) the emergence of the database as a framework for cultural and political thinking and the effects of datafication of the world.”, the Big Bang Data exhibition (Spain) (organised by the Centre de Cultura Contemporània de Barcelona (CCCB) and Fundación Telefónica 2015) took under consideration different topics about city’ shape and layout on the basis of different kind of data.

Academic literature based on Twitter-use patterns has developed at different geographical scales. The spatial pattern of social networks and the threshold of their activities have been analysed at urban scale across three cities in order to produce an inter-urban analysis (Bawa-Cavia 2010). Diverse methods, such as attempts to aggregate the activity on the network onto a grid (400 × 400 m) of dots representing the ‘walkable’ cells showed the threshold walking distances in urban areas (O’Sullivan and Morral 1996). At wider scale, human flows are gathered, extracted and analysed in geolocated social media data through different instruments. An example is FlowSampler, an interactive interface for visual analysis of flows in geolocated social media data. It adopts a graph-based approach to infer movement pathways from spatial point type data and expresses the resulting information through multiple linked visualizations to support data exploration (Chua et al. 2014). Additionally, it allows characterizing places in base the “density” of arc traces and keywords within the local groups. Big data also helps to study prescription planning such as to plan routes of buses and subways to satisfy public demand in several cities around the world (Fisher 2012).

Through the spatial analysis of UGD, we come to understand that some cities are far more active early in the morning, while others show higher activity at night or on weekends (Neuhaus 2011). The Centre for Advanced Spatial Analysis (CASA) at University College London (UCL) Twitter-use monitored several cities over a week to determine patterns in temporal activities and learn about spatial networks. Other analyses on UGD and available data from social networks - Twitter, Facebook, Flickr and Instagram - captured information about citizens’ feelings, citizens’ uses, citizens’ comfortability and conditions of problematic situations. Funded by the German Research Foundation, the “Urban Emotions” project (Heidelberg and Kaiserslautern Universities) tested UGD by checking whether the emotions measured correlate with the subjective assessments in the social media. “People as Sensors” used measures emotions and stress levels, such as “unsafe bike paths, traffic jam stress, frightening underpasses” (Heidelberg University 2014). Therefore, measurements of citizens’ emotional responses to their environment were made with by the use of sensors, similar to a wristwatch “that allows us to measure skin conductance, body temperature and variations in heart rate that change, for instance, when someone is startled” (Resch et al. 2015). Differently, but also tracing emotions, the exploration of how people express excitement online show a regular pattern; in which as much higher levels of excitement and more intense the flurry of messages in the collective, the shorter the messages become (Senseable City Lab and Ericsson 2015; Szell et al. 2014). Emotional bursts become faster and more impulsive online than offline. Many associated questions and outcomes become still unanswered, such as: “Are people doing this independently, or in response to seeing other short messages? Are we following the herd? Could we use these insights to learn more about financial bubbles by measuring more impulsive, less rational responses? And can we design better communication services?” (Lanzerotti et al. 2013).

Nevertheless, cautions have to be taken in considering the use of big data. There are many positive aspects on big data analysis, but many fears can be raised too. The obvious one is the intrusion on in the private sphere of our lives. In addition, putting data culture at the centre of decision-making and on the way of interpreting the world opens up many possibilities and involves numerous risks. The main danger of data-centrism is that it encourages the idea that whatever the problem, the answer lays in data. This important discussion escapes the goal of this chapter.

The several big data methodologies presented in this chapter are better understood if different geographical scales are used for analysis. Besides the metropolitan area of Barcelona, two very diverse morphological and social POS are chosen: Enric Granados and Forum (Fig. 1).

Enric Granados is within the consolidated historical city; and connects the Diagonal Avenue with a foundational building of the University of Barcelona. Differently from Rambla de Catalunya and Passeig de Gràcia, which are parallel to it, the Enric Granados width is slightly more than 20 m and hosts several leisure activities mostly for neighbours and local people. The urban transformations that followed the Olympic Games in 1992 changed the role of this street in the urban hierarchy. From a traditional car-oriented road, it became a pedestrian that is a user-oriented space. This was made possible by modifying the car mobility of the street.

The Forum rises on the old industrial Poblenou, an urban district that was renewed during the end of the twentieth, beginning of the twenty-one century; and hosted the Universal Forum of Cultures (2004) (Casellas, Dot-Jutgla and Pallares-Barbera 2012). Forum is where the Diagonal Avenue meets the Mediterranean Sea. The new character of the area is pointed by a number of new architectures, with building, skyscrapers designed by very famous architects. Nevertheless, the functional connectedness of the whole area is inexistent. The large paved esplanade provides a sense of incompleteness with no seats, nor trees, nor shadows.

The several methods used in this chapter are based on geo mapping. Focus is placed on intensity and concentration of users in open public spaces. Data was collected using 3 different ICT methodologies in 3 periods of time in Barcelona. First, Twitter data was collected by SiTI (Higher Institute on Territorial Systems for Innovation, Torino) during the period between January the 12th and 18th, 2015. Second, Twitter data was compiled using TSE (Tweeter Search Engine) application in January from the 13th to the 20th, 2017. Third, WAY app gathered data in November 27, 2014 (COST Action 1306 2014), with a selected group of users moving in and between the two case study areas.

Several spatial patterns have arisen from the planned methods analysing the use of Twitter and WAY app. Although these initial analyses can be developed further, their opportunities for specific additional improvements are open in future research; where data comparisons, fitness of the methodology and so on are to be implemented to meet diverse research structural questions.

The overview on the state of the art shows how spatial planning is facing a transition presenting new challenging scenarios. The trend led by the smart city opened the spatial planning to logic of numbers and data. City-sensors as well as UGC systems constantly provide huge amounts of data which can be used to supply the citizens’ demands. In this context, two main branches are emerging. The first branch is a data-oriented and technology-driven approach, which makes use of quantities for assessing and justifying decisions. Eye-catching visualizations are the most evident outputs of such an approach, where analysts and statistics prevail on the human experience of professionals and experts.

The second branch uses new technologies to improve human abilities and it is mostly used to support the decision and policy-making processes. Particular efforts are spent in understanding how data can support can be complementary to the traditional approach, providing new insights on spatial issues. The debate on the use of big data is nowadays shifting from a technology-driven vision towards a more human dimension, introducing the concepts of people friendliness and a human-to-human approach (Melis et al. 2014). To achieve this social vision, the collaboration between different experts such as urban planners and data analysts and designers appears essential in order to guide them towards relevant questions and policy issues (Goodspeed 2012). Simple visualizations offer opportunity for both horizontal and vertical communication.

From a technical point of view, some considerations are necessary for improving future map production. First of all, paths created using subsequent tweets are just possible routes and probably not the real ones covered by users. A method for dealing with this issue is necessary, and different research is being undertaking. Secondly, the use of Twitter data is most suitable for larger scale than the micro-urban areas. However, the continuity along space and over time provide important element for considering their hierarchical position.

In this chapter, the analysis was limited to visualisation and maps; it did not considered semantics, the written content of the tweets (however, a method for experiencing has been suggested); statistical testing has not been performed. Nevertheless, the humble objectives have been stated and proved. We know that integration of new methodologies within spatial planning processes are very relevant, such as in the data collecting and information; defining goals, objectives and strategies and implementation of users’ participation through crowdsourcing.