Abstract
To hold software service and platform providers accountable, it is necessary to create trustworthy, quantified evidence of problematic algorithmic decisions, e.g., by large-scale black box analyses. In this article, we summarize typical and general challenges that arise when such studies are conducted. Those challenges were encountered in multiple black box analyses we conducted, among others in a recent study to quantify, whether Google searches result in search results and ads for unproven stem cell therapies when patients research their disease and possible therapies online. We characterize the challenges by the approach to the black box analysis, and summarize some of the lessons we learned and solutions, that will generalize well to all kinds of large-scale black box analyses. While the studies we base this article on where one-time studies with an explorative character, we conclude the article with some challenges and open questions that need to be solved to hold software service and platform providers accountable with the help of permanent, large-scale black box analyses.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
These initial impressions were collected during the Wellcome Trust Seed project-funded workshop “Patienthood and Participation in the Digital Era: findings and future directions” hosted by the Usher Institute at the University of Edinburgh in August 2018. (Erikainen et al. [14]).
- 2.
References
Andreou, A., Venkatadri, G., Goga, O., Gummadi, K., Loiseau, P., Mislove, A.: Investigating Ad transparency mechanisms in social media: a case study of Facebook’s explanations. In: NDSS 2018 - Network and Distributed System Security Symposium, San Diego, CA, United States, pp. 1–15 (2018)
Ashby, W.R.: An Introduction to Cybernetics. Chapman & Hall Ltd., London (1957)
Biddings, A.: A new policy on advertising for speculative and experimental medical treatments. Google Ads Help (2019). https://support.google.com/google-ads/answer/9475042. Accessed 11 Mar 2021
Bovens, M.: Analysing and assessing accountability: a conceptual framework1. Eur. Law J. 13(4), 447–468 (2007). https://doi.org/10.1111/j.1468-0386.2007.00378.x
Bucher, T.: Neither black nor box: ways of knowing algorithms. In: Kubitschko, S., Kaun, A. (eds.) Innovative Methods in Media and Communication Research, pp. 81–98. Palgrave Macmillan, Cham (2016). https://doi.org/10.1007/978-3-319-40700-5_5
Cadwalladr, C.: Facebook’s role in Brexit - and the threat to democracy (2019). TED Talk. https://www.ted.com/talks/carole_cadwalladr_facebook_s_role_in_brexit_and_the_threat_to_democracy. Accessed Mar 11 2021
Datta, A., Tschantz, M.C., Datta, A.: Automated experiments on Ad privacy settings. Proc. Priv. Enhancing Technol. 2015(1), 92–112 (2015). https://doi.org/10.1515/popets-2015-0007
DEK: Gutachten der Datenethikkommission der deutschen Bundesregierung. Bundesministerium des Innern, für Bau und Heimat (2019). https://www.bmi.bund.de/SharedDocs/downloads/DE/publikationen/themen/it-digitalpolitik/gutachten-datenethikkommission.pdf. Accessed 11 Mar 2021
Diakopoulos, N.: Algorithmic accountability reporting: on the investigation of black boxes, tow center for digital journalism/knight brief. Columbia Journal. School (2014). https://doi.org/10.7916/D8ZK5TW2
Diakopoulos, N.: Algorithmic accountability. Journalistic investigation of computational power structures. Digit. Journal. 3(3), 398–415 (2015). https://doi.org/10.1080/21670811.2014.976411
Dixon, E., Enos, E., Brodmerkle, S.: A/B testing of a webpage. United States Patent (2011). https://patentimages.storage.googleapis.com/35/bf/a3/2a1ee861e2adaf/US7975000.pdf. Accessed 11 Mar 2021
Enquete-Kommission: Künstliche Intelligenz – GesellschaftlicheVerantwortung und wirtschaftliche, soziale und ökologische Potenzialedes Deutschen Bundestags: Abschlussbericht. Berlin, Drucksache 19/23700 (2020). Accessed 11 Mar 2021
Erikainen, S., Couturier, A., Chan, S.: Marketing experimental stem cell therapies in the UK: biomedical lifestyle products and the promise of regenerative medicine in the digital era. Sci. Cult. 29(2), 219–244 (2020). https://doi.org/10.1080/09505431.2019.1656183
Erikainen, S., Pickersgill, M., Cunningham-Burley, S., Chan, S.: Patienthood and participation in the digital era. Digit. Health (2019). https://doi.org/10.1177/2055207619845546
Europäische Kommission: Weißbuch Zur Künstlichen Intelligenz - ein europäisches Konzept für Exzellenz und Vertrauen (2020). Europäische Kommission https://ec.europa.eu/info/sites/info/files/commission-white-paper-artificial-intelligence-feb2020_de.pdf. Accessed 11 Mar 2021
Gillespie, T.: The relevance of algorithms. In: Media Technologies: Essays on Communication, Materiality, and Society, p. 167 (2014)
Google: Healthcare and medicines. Google advertising policies help (2019). https://support.google.com/adspolicy/answer/176031. Accessed 11 Mar 2021
Granka, L.A.: The politics of search: a decade retrospective. Inf. Soc. 26(5), 364–374 (2010). https://doi.org/10.1080/01972243.2010.511560
Grunwald, A.: Technikfolgenabschätzung - Eine Einführung. Edition Sigma, Berlin (2002)
Introna, L.D.: Algorithms, governance, and governmentality: on governing academic writing. Sci. Technol. Human Values 41(1), 17–49 (2016). https://doi.org/10.1177/0162243915587360
Kienle, A.: Integration von Wissensmanagement und kollaborativem Lernen durch technisch unterstützte Kommunikationsprozesse. Dissertation, Universität Dortmund, Dortmund (2003)
Kitchin, R.: Thinking critically about and researching algorithms. Inf. Commun. Soc. 20(1), 14–29 (2017). https://doi.org/10.1080/1369118X.2016.1154087
Krafft, T.D., Gamer, M., Zweig, K.A.: What did you see? A study to measure personalization in Google’s search engine. EPJ Data Sci. 8(1), 38 (2019)
Krafft, T.D., Hauer, M.P., Zweig, K.A.: Why do we need to be bots? What prevents society from detecting biases in recommendation systems. In: Boratto, L., Faralli, S., Marras, M., Stilo, G. (eds.) BIAS 2020. CCIS, vol. 1245, pp. 27–34. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-52485-2_3
Noble, S.U.: Algorithms of Oppression - How Search Engines Reinforce Racism. New York United Press, New York (2018)
Pagano, G., Ferrara, N., Brooks, D.J., Pavese, N.: Age at onset and Parkinson disease phenotype. Neurology 86(15), 1400–1407 (2016). https://doi.org/10.1212/WNL.0000000000002461
Prainsack, B.: Data donation: how to resist the iLeviathan. In: Krutzinna, J., Floridi, L. (eds.) The Ethics of Medical Data Donation. PSS, vol. 137, pp. 9–22. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-04363-6_2
Reber, M., Krafft, T.D., Krafft, R., Zweig, K.A., Couturier, A.: Data donations for mapping risk in google search of health queries: a case study of unproven stem cell treatments in SEM. In: IEEE Symposium Series on Computational Intelligence (SSCI), pp. 2985–2992 (2020)
Sandvig, C., Hamilton, K., Karahalios, K., Langbort, C.: Auditing algorithms: research methods for detecting discrimination on internet platforms. Data Discrim. Conv. Crit. Concerns Prod. 22, 4349–4357 (2014)
Seaver, N.: Knowing algorithms. In: Media in Transition, Cambridge, MA, vol. 8 (2014)
Wieringa, M.: What to account for when accounting for algorithms: a systematic literature review on algorithmic accountability. In: Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency, pp. 1–18 (2020)
Zweig, K.A., Krafft, T.D., Klingel, A., Park, E.: Sozioinformatik Ein neuer Blick auf Informatik und Gesellschaft. Carl Hanser Verlag (2021, in publication)
Acknowledgment
The presented project EDD has been partially funded by the EU stem cell public engagement project, EuroStemCellFootnote 2 and by a generous grant from the University of Edinburgh School of Social and Political Science. The research was supported by the project GOAL “Governance of and by algorithms (Funding code 01IS19020) which is funded by the German Federal Ministry of Education and Research.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Krafft, T.D., Reber, M., Krafft, R., Coutrier, A., Zweig, K.A. (2021). Crucial Challenges in Large-Scale Black Box Analyses. In: Boratto, L., Faralli, S., Marras, M., Stilo, G. (eds) Advances in Bias and Fairness in Information Retrieval. BIAS 2021. Communications in Computer and Information Science, vol 1418. Springer, Cham. https://doi.org/10.1007/978-3-030-78818-6_13
Download citation
DOI: https://doi.org/10.1007/978-3-030-78818-6_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-78817-9
Online ISBN: 978-3-030-78818-6
eBook Packages: Computer ScienceComputer Science (R0)