Developing a privacy-aware map-based cross-platform social media dashboard for municipal decision-making

Introduction

Users of location-based social media networks (LBSN), such as Instagram,
Flickr, or Twitter, have produced an unprecedented base of data over the
past decade. According to ILIEVA & MCPHEARSON (2018: 553), "the
enormous scale and timely observation are unique advantages of [social
media data]" and therefore hold enormous potential for various
application purposes such as urban planning, among others.

Most notably for Instagram, as one of the largest LBSN, encouraging the
sharing of locations when creating content, offers completely new and
promising application purposes, through the combination of the spatial
component with timestamps and the actual content (image & text).

Public social media (SM) data have shown their potential examining the
increasingly relevant social problems of Spatial (In-) Justice, spatial
(in-) equality and spatial (in-) equity (Cf. SOJA 2013: 47). However,
few research attempts were made to make these results available broader
in practice and accessible to laypersons in an understandable way.

LBSN data could contribute significantly to creating a better
information base for municipal decision-making processes, reaching
especially younger target groups. Until now, specifically these groups
were difficult to reach in common participation processes (Cf. SELLE
2004), while bearing consequences of municipal policies for the longest
period of time.

Our stated research goal is therefore to provide citizens, laypersons
and municipal decision-makers with an unprecedented LBSN Dashboard, as a
simple open-source platform for spatial multi-purpose LBSN analysis.

Problem Statement

Such an undertaking raises certain ethical and legal questions, since
the user data belong to the users themselves, including the right to
self-determination over their data, on the one hand, and the right to
privacy on the other. The far too short-sighted (but frequently used)
argument that posts have been deliberately published, with all the
consequences of their public nature in mind (e.g., BURTON et al. 2012:
2), is simply not sufficient for an in-depth discussion of privacy. This
further violates the most important aspects of privacy (Cf. BOYD &
CRAWFORD 2012: 672). In fact, most users are not or only partially aware
of what can actually be inferred from what they share or disclose about
themselves (KESSLER & MCKENZIE 2018: 6f).

Yet, privacy is rarely addressed in LBSN research and, worse, often
negligently ignored. In this context, many negative examples can be
found where data was analyzed and high-resolution results were
published, clearly violating users' privacy, for example, in scientific
publications (Cf. KOUNADI & LEITNER 2014: 140).

Research Interest

Given the increasing socio-spatial inequality, the rapid growth of SM,
and the growing interest of municipalities in SM knowledge, we see a
significant need for such a privacy-aware LBSN dashboard, which is
entirely new to the geospatial community.

We develop a privacy-aware LBSN dashboard prototype and propose a data
processing pipeline based on the HyperLogLog (HLL) algorithm by FLAJOLET
et al. (2007). The dashboard is geared towards easy information
retrieval and making use of the data richness of LBSN -- without
compromising user privacy and the need for extensive data retention.
Instead, we provide a unique, customizable, GDPR-compliant privacy
approach. The combination of different open-source tools for structuring
multi-platform LBSN data, leveraging the capabilities of HyperLogLog and
simple Python integration ensure easy reproducibility and active
community development (Cf. DUNKEL et al. 2021; DUNKEL & LÖCHNER 2021a &
b).

The dashboard prototype is tailored for use in municipalities and its
citizens, but offers high scalability for other purposes or other
spatial levels. A limited interactive demo and its GitHub repository are
permanently publicly available as a result of a Master's thesis and an
IoT Design Thinking Workshop (Cf. WECKMÜLLER 2021; BUNDESSTADT BONN
2022).

We plan on finishing and automatizing the data processing pipeline,
enabling more sophisticated queries and adding further visualization
methods. In the long run, the dashboard is thought to serve as a
participation and open data hub for all citizens and for any city in the
world. So far, the city of Bonn and Chemnitz (Germany) are pilot
partners of this research project.

References

Literature

BOYD, D., & CRAWFORD, K. (2012). CRITICAL QUESTIONS FOR BIG DATA.
Information, Communication & Society, 15(5), 662--679.

BURTON, S. H., TANNER, K. W., GIRAUD-CARRIER, C.G., WEST,J. H., &
BARNES, M. D. (2012). "Right Time, Right Place" Health Communication
on Twitter: Value and Accuracy of Location Information. Journal of
medical Internet research, 14(6), e156.

FISCHER, F. (2008). Location Based Social Media -- Considering the
Impact of Sharing Geographic Information on Individual Spatial
Experience. In A. Car, G. Griesebner, & J. Strobl (Eds.) Geospatial
Crossroads @ GI_Forum '08. Proceedings of the Geoinformatics Forum
Salzburg (pp. 1-7). Wichmann.

FLAJOLET, P., FUSY, É., GANDOUET, O., & MEUNIER, F. (2007). Hyperloglog:
the analysis of a nearoptimal cardinality estimation algorithm. Analysis
of Algorithms 2007 (AofA07), 127--146.

ILIEVA, R. T., & MCPHEARSON, T. (2018). Social-media data for urban
sustainability. Nature Sustainability, 1(10), 553-565.

KESSLER, C., & McKenzie, G. (2018). A geoprivacy manifesto. Transactions
in GIS, 22(1), 3-19.

KOUNADI, O., & LEITNER, M. (2014). Why does geoprivacy matter? The
scientific publication of confidential data presented on maps. Journal
of Empirical Research on Human Research Ethics, 9(4), 34-45.

Selle, K. (2004). Kommunikation in der Kritik? In: Müller B., Löb S.,
Zimmermann K. (Ed.) Steuerung und Planung im Wandel, VS Verlag für
Sozialwissenschaften.

SOJA, E. W. (2013). Seeking spatial justice (Vol. 16). University of
Minnesota Press.

List of Web References

All links last accessed on February 20, 2022.

BUNDESSTADT BONN (2022). Studierende entwickeln neue Ideen für die
digitale Stadt von morgen.
https://www.bonn.de/pressemitteilungen/januar-2022/studierende-entwickeln-neue-ideen-fuer-die-digitale-stadt-von-morgen.php

DUNKEL, A., LÖCHNER, M., KRUMPE, F. & Contributors (2021). LBSN
Structure. https://lbsn.vgiscience.org/.

DUNKEL, A. & LÖCHNER M. (2021a). LBSN HLL Database - Docker Container.

https://gitlab.vgiscience.de/lbsn/databases/hlldb/

DUNKEL, A. & LÖCHNER M. (2021b). Lbsntransform.
https://lbsn.vgiscience.org/lbsntransform/docs/.

WECKMÜLLER, D. (2021). LBSN-Dashboard Prototype for Bonn.
https://geo.rocks/lbsndashboard/