Revealing the determinants of gender inequality in urban cycling with large-scale data, by A. Battiston, L. Napoli, P. Bajardi, A. Panisson, A. Perotti, M. Szell & R. Schifanella, published in EPJ Data Science

The uptake of cycling in today’s cities is especially low for women: there is a largely unexplained, persistent gender gap in cycling. To understand the determinants of this gender gap in cycling at scale, here we use massive, automatically-collected data from the tracking application Strava on outdoor cycling for 61 cities across the United States, the United Kingdom, Italy and the Benelux area. While Strava data is particularly well-suited to describe the behavior of regular cyclists and its generalizability to occasional cyclists requires further investigation, the size of these data and their characteristics represent an unprecedented opportunity for the literature on cycling. Leveraging the associated gender and usage information, we first quantify the emerging gender gap in recreational cycling at city-level. A comparison of cycling rates of women across cities within similar geographical areas—where the penetration of Strava is assumed to be comparable—unveils a broad range of gender gaps. On a macroscopic level, we link this heterogeneity to a variety of urban indicators and provide evidence for traditional hypotheses on the determinants of the gender-cycling-gap. We find a positive association between female cycling rate and urban road safety. On a microscopic level, we identify female preferences for street-specific features in the city of New York. Assuming that the determinants of the gender-cycling-gap are similar across regular and occasional cyclists, our study suggests that enhancing the quality of the dedicated cycling infrastructure may be a way to make urban environments more accessible for women, thereby making urban transport more sustainable for everyone.

Today we published a paper on ideological polarization. Special congrats to Marilena for this being her first paper!

An intensely debated topic is whether political polarization on social media is on the rise. We can investigate this question only if we can quantify polarization, by taking into account how extreme the opinions of the people are, how much they organize into echo chambers, and how these echo chambers organize in the network. Current polarization estimates are insensitive to at least one of these factors: they cannot conclusively clarify the opening question. Here, we propose a measure of ideological polarization which can capture the factors we listed. The measure is based on the Generalized Euclidean (GE) distance, which estimates the distance between two vectors on a network, e.g., representing people’s opinion. This measure can fill the methodological gap left by the state of the art, and leads to useful insights when applied to real-world debates happening on social media and to data from the US Congress.

Multidimensional tie strength and economic development, by L.M. Aiello, S. Joglekar, and D. Quercia, published in Scientific Reports

For decades, Granovetter’s tie strength has been quantified using the frequency of interaction. Yet, frequency does not reflect Granovetter’s initial conception of strength, which is a mix of social dimensions including exchnage of knowledge and provision of support. We used Natural Language Processing to quantify whether text messages convey expressions of knowledge or support, and applied it to a large conversation network from of Reddit users resident in the United States. Borrowing a classic experimental setup, we tested whether the diversity of social connections of Reddit users resident in a specific US state would correlate with the economic opportunities in that state (estimated with GDP per capita). We found that the combination of diversity calculated on the knowledge and support networks correlates much more strongly with GDP than diversity calculated on a network weighted with interaction frequency (R2=0.62 vs. R2=0.30). We also found that the two types of ties differ in their geographical span. Knowledge ties are long-distance (i.e., connecting people living in far-away states), support ties are mostly created among people living close by. Read more in this blogpost.

We published another urban planning paper:

1. Computational Desire Line Analysis of Cyclists on the Dybbølsbro Intersection in Copenhagen, by S.M. Breum, B. Kostic, and M. Szell, published in Transport Findings

   

   To improve intersection planning, here we develop a computational method to detect cyclist trajectories from video recordings and apply it to the Dybbølsbro intersection in Copenhagen, Denmark. In one hour of footage we find hundreds of trajectories that contradict the design, explainable by the desire for straightforward, uninterrupted travel largely not provided by the intersection. This neglect and the prioritization of vehicular traffic highlight opportunities for improving Danish intersection design.

We are on an urban planning streak, publishing two new papers in Environment and Planning B:

1. Urban form and COVID-19 cases and deaths in Greater London: An urban morphometric approach, by A. Venerandi, L.M. Aiello, and S. Porta, published in Environment and Planning B
   
   The COVID-19 pandemic generated a considerable debate in relation to urban density. Many urban planners have advocated for rethinking our cities in ways that can decrease built-up density, in order to curb the spreading of future epidemics. In this work, we show that would be a bad idea. We used urban morphometrics to quantify built-up density in Greater London, and studied its relationship with COVID-19 cases and deaths at the level of MSOAs (small neighborhoods with an average population of ~8000). We found that urban density weakly and negatively correlates with both deaths and cases. The picture above (the low-density areas that some think could save us from contagion) shows the typical area in London with highest occurrence of COVID cases. The widespread belief that COVID cases scale with built-up density was supported mostly by city-level studies. The picture changes when comparing different areas within a city, which has been done for the first time in our study. The moral of the story is that built-up density is different from crowding. Let’s keep that in mind before worsening the urban sprawl of our cities. 
2. Data-driven micromobility network planning for demand and safety, by P. Folco, L. Gauvin, M. Tizzoni, and M. Szell, published in Environment and Planning B
   
   In this paper we study how data of micromobility trips and crashes can shape and automatize infrastructure network planning processes. We introduce a parameter that tunes the focus between demand-based and safety-based development, and investigate systematically this tradeoff for the city of Turin. We find that a full focus on demand or safety generates different network extensions in the short term, with an optimal tradeoff in-between. In the long term, our framework improves overall network quality independent of short-term focus. Thus, we show how a data-driven process can provide urban planners with automated assistance for variable short-term scenario planning while maintaining the long-term goal of a sustainable, city-spanning micromobility network.
   See the interactive visualization: http://www.datainterfaces.org/projects/biketracks/#turin

We have a new NERDS paper out by our Postdoc Sandro Sousa. This paper marks his main work from his PhD at Queen Mary University of London, published now:

1. Quantifying ethnic segregation in cities through random walks by S. Sousa and V. Nicosia, published in Nature Communications
   
   We propose here a family of non-parametric measures for spatial distributions, based on the statistics of the trajectories of random walks on graphs associated to a spatial system. These quantities provide a consistent estimation of segregation in synthetic spatial patterns, and we use them to analyse the ethnic segregation of metropolitan areas in the US and the UK. We show that the spatial diversity of ethnic distributions, as measured through diffusion on graphs, allow us to compare the ethnic segregation of urban areas having different size, shape, or peculiar microscopic characteristics, and exhibits a strong association with socio-economic deprivation.

Two new NERDS papers are out – this time both in Scientific Reports:

1. Heterogeneous rarity patterns drive price dynamics in NFT collections, by A. Mekacher, A. Bracci, M. Nadini, M. Martino, L. Alessandretti, L.M. Aiello & A. Baronchelli, published in Scientific ReportsWe quantify Non Fungible Token (NFT) rarity and investigate how it impacts market behaviour. We show that, on average, rarer NFTs: (i) sell for higher prices, (ii) are traded less frequently, (iii) guarantee higher returns on investment, and (iv) are less risky, i.e., less prone to yield negative returns. The dataset used for the work has been presented as part of a beautiful art exhibition at the MEET Digital Culture Center in Milano.
2. Posts on central websites need less originality to be noticed, by M. Coscia and C. Vandeweerdt, published in Scientific Reports
   In this paper we study how originality and centrality interact in a nontrivial way, which might explain why originality by itself is not a good predictor of success. We collected data from Reddit on users sharing hyperlinks. We estimated the originality of each post title and the centrality of the website hosting the shared link. We show that the interaction effect exists: if users share content from a central website, originality no longer increases the odds of receiving at least one upvote.
   See more info in Michele’s blogpost: https://www.michelecoscia.com/?p=2205

We published two more papers over the last weeks!

1. 1. A potential mechanism for low tolerance feedback loops in social media flagging systems, by C.J. Westermann and M. Coscia, published in PLOS ONE

      In this paper we simulate a scenario in which users on one side of the polarity spectrum have different tolerance levels for the opinions of the other side. We create a model based on some assumptions about online news consumption, including echo chambers, selective exposure, and confirmation bias. When studying a model of social media flagging, we see that intolerance is attractive: news sources are nudged to move their polarity to the side of the intolerant users.
      See more info on Michele’s blogpost: https://www.michelecoscia.com/?p=2179

   2. Multimodal urban mobility and multilayer transport networks, by L. Alessandretti, L.G. Natera Orozco, M. Saberi, M. Szell, F. Battiston, published in Environment and Planning B
      

      This is a comprehensive overview of the emerging research areas of multilayer transport networks and multimodal urban mobility, focusing on contributions from the interdisciplinary fields of complex systems, urban data science, and science of cities. First, we present an introduction to the mathematical framework of multilayer networks. We apply it to survey models of multimodal infrastructures, as well as measures used for quantifying multimodality, and related empirical findings. We review modelling approaches and observational evidence in multimodal mobility and public transport system dynamics, focusing on integrated real-world mobility patterns, where individuals navigate urban systems using different transport modes. We then provide a survey of freely available datasets on multimodal infrastructure and mobility, and a list of open source tools for their analyses. Finally, we conclude with an outlook on open research questions and promising directions for future research.

We just published two more papers! Both are on the topic of sustainable mobility:

1. 1. Growing urban bicycle networks, by M. Szell, S. Mimar, T. Perlman, G. Ghoshal, and R. Sinatra, published in Scientific Reports
      

      Here we explore systematically the topological limitations of urban bicycle network development. For 62 cities we study different variations of growing a synthetic bicycle network between an arbitrary set of points routed on the urban street network. We find initially decreasing returns on investment until a critical threshold, posing fundamental consequences to sustainable urban planning: Cities must invest into bicycle networks with the right growth strategy, and persistently, to surpass a critical mass. Growing networks from scratch makes our approach a generally applicable starting point for sustainable urban bicycle network planning with minimal data requirements.
      The paper comes with an accompanying data visualization: https://growbike.net

   2. Identifying urban features for vulnerable road user safety in Europe, by M. Klanjcic, L. Gauvin, M. Tizzoni, and M. Szell, published in EPJ Data Science

      We identify urban features that are determinants of vulnerable road user safety through the analysis of inter-mode collision data across 24 European cities. We observe that cities with the highest rates of walking and cycling modal shares are the safest for the most vulnerable users. Our results suggest that policies aimed at increasing the modal share of walking and cycling are key to improve road safety for all road users.
      We explain and motivate our project in this accompanying blogpost (to appear on https://blogs.biomedcentral.com): 

      Identifying urban features for vulnerable road user safety in Europe

      Which traffic participants and which urban features are most associated with road deaths? Recently published work in EPJ Data Science explores this question with data from 24 European cities.

      

      Road crashes result in yearly 1.3 million deaths and 2.3 trillion USD of economic damage. Because of this pressing societal issue, the UN has declared in 2015 the global sustainability goal to halve the number of road deaths by 2020. This goal failed: Road deaths keep rising worldwide.

      Many cities are wondering how to solve this issue. However, they might not have the full picture because road crash statistics tend to be reported in a victim-centered way: There are detailed statistics on the distributions of victim demographics such as age or gender, but this neglects necessary information for answering two important questions towards better crash prevention: 1) Who causes the crashes? 2) Why do crashes happen?

      The first question can be explored via the so-called casualty matrix. It shows the casualties between all combinations of different traffic participants, for example the threat of cars on cars, of cars on pedestrians, or the threat of trucks on cyclists.

      As has been previously shown in an impressive data visualization by a Dutch journalist and researcher team, the by far biggest threat to human life on urban streets in the Netherlands is motorized vehicles – cars and trucks – while cyclists and pedestrians are overwhelmingly their victims and harmless. This sounds plausible, but a systematic, quantitative study over multiple countries has been missing.

      The second question – Why do crashes happen? – is much harder to answer. Generally crashes happen in an interplay between the individual behaviors of crash participants and their environments.

      Environmental features like the extent of pedestrian areas, cycling tracks, or speed limits, are easier to collect than behavioral data, therefore their relation to road user risk could be explored in a straightforward way. And because the environment can be changed or regulated by decision makers, they can be held responsible to act.

      To support with evidence such actions towards making cities safer, the OECD recently called for developing a modern approach to road safety: 1) collect and analyze crash data from a larger set of cities, 2) investigate the relationships between urban shape, density, speeds, and road user risk, and 3) analyze casualty matrices.

      Inspired by these OECD recommendations and the Dutch data visualization, in our work we collected crash data from 24 cities in 5 European countries with high enough resolution to build and explore casualty matrices, to quantify road safety in a systemic approach, and to identify those urban features that are most relevant, especially for vulnerable road users like pedestrians who are known to be disproportionally impacted.

      Exploring the casualty matrices first, we found the same overall picture as our Dutch colleagues, see Fig. 1: Cars are the most substantial hazard. However, we also found considerable local variations. For example, cars are a considerable threat to pedestrians and cyclists in inner London, but this is much less the case in Barcelona.

      Figure 1: Casualty matrices for 2018 show road deaths and serious injuries after a traffic participant on the left collided with one on the bottom. Cars are responsible for the majority of road deaths/injuries, while columns for pedestrians and cyclists do not appear because they pose practically no risk to others.

      When normalizing the number of deaths and serious injuries by population, we found British cities to be most dangerous, while Oslo is by far the least dangerous. This is not surprising given how much Oslo has recently invested into following a Vision Zero strategy, i.e. to aim for zero road fatalities, which they achieved in 2019.

      Finally, we set out to answer: What are the urban features most associated with crashes? Here we considered several features acquired from different sources, like OpenStreetMap: population density, the amount of bicycle tracks versus car lanes, the fraction of low-speed-limit streets, the distribution of how people move (walking, cycling, public transport, or motor vehicles), temperature, precipitation, and GDP.

      Using an information theory measure to identify the most fitting pairings of these features with road crashes, we found the best significant predictor, see Fig. 2: Cities with more people walking have less road deaths.

      Figure 2: The share of people walking in a city is a significant predictor for less casualties, for any traffic participant killed or seriously injured by a car. Numbers are regression coefficients, black borders denote statistical significance at p < 0.05.

      Interestingly, this result extends to all modes of transport: More walking is associated not just with higher pedestrian safety, but also with higher cyclist and motorist safety.

      Apart from the share of walking, a similarly strong association with road safety is having more streets limited to at most 30 km/h.

      We need to be clear that our results can be only as good as the underlying data, and these can have large reporting biases. For example, crashes with cyclists are often not reported, and different EU countries have different reporting procedures. So, more research and more standardized policies on crash reporting are needed. Also, we only calculated statistical correlations, so we cannot say anything about cause and effect.

      Nevertheless, our data-driven conclusions support a modern, evidence-based paradigm of road safety, suggesting this advice to urban decision makers: Make your cities more walkable and remove the hazard of cars. Besides the massive public health benefits, this will make your city more livable and its transport system more sustainable.

We are on a streak and just published two more papers!

1. 1. From Reddit to Wall Street: the role of committed minorities in financial collective action, by L. Lucchini, L.M. Aiello, L. Alessandretti, G. De Francisci Morales, M. Starnini and A. Baronchelli, published in Royal Society Open Science
      
      We analyzed the coordinated activity on Reddit that led to target short-selling activity by hedge funds on GameStop shares, causing a surge in the share price and triggering significant losses for the funds involved. We found that a small fraction of individuals can trigger large behavioural cascades, and we show the role of commitment, network centrality, and social identity in such coordination process. Our findings shed light on financial collective action, which several observers anticipate will grow in importance.
   2. Generalized Euclidean Measure to Estimate Distances on Multilayer Networks, by M. Coscia, published in ACM Transactions on Knowledge Discovery from Data
      
      In this paper we propose an algorithm solving node vector distance for multilayer networks, which is a problem that arises for all kinds of network spreading processes like epidemics, economic growth, or human behavior. We do so by adapting the Mahalanobis distance, incorporating the graph’s topology via the pseudoinverse of its Laplacian. Since this is a proper generalization of the Euclidean distance in a complex space defined by the topology of the graph, and that it works on multilayer networks, we call our measure the Multi Layer Generalized Euclidean (MLGE).