The paper looks into which apps people use and creates app-fingerprints for 3.5 million individuals. Similar to forensic science where you need 12 points to distinguish between fingerprints we ask how many apps do we need to distinguish between two users? We find people’s smartphone app behavior is very unique and 3 apps are enough to identify more than 90% of all individuals. But app-fingerprints change over time and are different between countries. We find that people have more unique app-fingerprints during summer because we use more unique apps, and Americans have the most unique fingerprints (need the fewest apps to identify them) while Finns are the least unique (need more apps to identify their fingerprint). Why is this important? Because the work highlights problems with current policies intended to protect user privacy and emphasizes that policies cannot directly be ported between countries.
With NERDS winning grants and growing, we have added a Positions page at https://nerds.itu.dk/positions/, currently featuring 2 open calls at ITU that are directly relevant to us as they can lead to NERDS positions. If your research overlaps with ours and you are interested get in touch!
1) PhD Open Call 2021
The ITU-wide PhD Open Call 2021, deadline March 10th, features 2 potential PhD projects by NERDS members:
1) Michael: Network analysis of urban transport networks for a green transition from car- centricity to cycling
2) Michele+Luca: Modelling Complex Social Systems to Handle Disinformation
2) Asst./Assoc. Professor in data science and machine learning
On yesterday’s International Day of Women and Girls in Science, Roberta took part in a corresponding event organized by her alma mater, University of Catania, Dipartimento di Fisica e Astronomia “Ettore Majorana”, delivering a keynote talk on the topic and on her research.
We are thrilled to welcome Luca Maria Aiello to our research group!
Luca joins us as Associate Professor, coming from industry. He conducts interdisciplinary research at the intersection of computational social science, digital health, network science, and urban informatics, using large-scale digital data to quantify people’s well-being and build systems that can improve it. Currently, he is focusing on Natural Language Processing to quantify social and psychological experiences from text.
He had a few past professional roles: Senior Research Scientist at Bell Labs in Cambridge, UK; Research Fellow of the ISI Foundation in Turin; Research Scientist at Yahoo Labs Barcelona and London; visiting scientist at the Center for Complex Networks and Systems at Indiana University.
The Villum Young Investigator programme (YIP) focuses on attracting and retaining talented young Danish and international researchers at Danish universities. The aim is to support the development of high-level international research environments in the universities.
Roberta’s winning proposal was awarded with DKK 6M: Bias Explained: Pushing Algorithmic Fairness with Models and Experiments
Algorithms for ranking scientific information have an issue: they use citations, which are ingrained with human biases. Therefore, their output is also biased, creating inequalities and raising concerns of discrimination. This project aims to uncover the mathematical bias mechanisms that drive different citation trajectories given same quality, and to use them for creating fair algorithms.
We are overwhelmed with joy for Roberta’s success, and are looking forward to her future groundbreaking research. The grant will allow the recruitment of one PhD student and two postdocs – so stay tuned for upcoming job calls.
In the past two years, NERDS member Michele Coscia has been working on a textbook for the Network Analysis class he teaches at ITU. This “Atlas for the Aspiring Network Scientist”, has now reached version 1.0, and 760 pages, and is available for anyone to read for free: https://arxiv.org/abs/2101.00863
The Atlas aims at being broad, not deep, to be a pointer to the things you need to know about Network Science rather than a deep explanation of those things.
Consider this a v1.0 of a continuous effort. There are many things to improve: language, concepts, references, figures. Please contact Michele with any comments.
Michele also plans to have interactive figures on the website in the future. Version 1.0 was all financed using his research money and time. For the future, Michele will need some support to do this in his free time. If you feel like encouraging this effort, you can consider becoming a member on Patreon.
Multilayer networks are an increasingly popular way to model complex relations between various types of entities and they have been applied to a large number of real-world data sets. Their intrinsic complexity makes the visualization of this type of network extremely challenging and still an open research area. To help the visual exploration of complex multilayer network structures, today we are releasing MNET-VR. MNET-VR is the output of a research project carried on with Leonard Maxim and supported by the Digital Design Department. MNET-VR explores the potential of Virtual Reality to visualize this type of network structure.
MNET-VR offers basic functions to visualize and filter multilayer network structures. MNET-VR does not offer, at this stage, the possibility to manipulate the network layout. A proper 3D layout of the network can be obtained through the R package multinet. While its primary goal is to explore multilayer networks, MNET-VR can also be used to visualize single layer networks using igraph and multinet. The export of of the files for visualization is done through two simple R functions that we make available on the website. MNET-VR is designed for Oculus Rift/Oculus Quest with Link.
Data-driven strategies for optimal bicycle network growth, by L.G. Natera Orozco, F. Battiston, G. Iniguez, M. Szell, published in Royal Society Open Science
Here we investigate the network structure of bicycle networks in cities around the world, and find that they consist of hundreds of disconnected patches, even in cycling-friendly cities like Copenhagen. To connect these patches, we develop and apply data-driven, algorithmic network growth strategies, showing that small but focused investments allow to significantly increase the connectedness and directness of urban bicycle networks.