scholarly journals ALGORITHMIC COPYRIGHT ENFORCEMENT ON YOUTUBE: USING MACHINE LEARNING TO UNDERSTAND AUTOMATED DECISION-MAKING AT SCALE

2019 ◽  
Vol 2019 ◽  
Author(s):  
Joanne Gray ◽  
Nicolas Suzor
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Digital Media ◽  
Large Scale ◽  
Multinomial Logistic Regression ◽  
Policy Research ◽  
Neural Nets ◽  
Automated Systems ◽  
Copyright Enforcement ◽  
Automated Decision Making

This paper presents the results of an investigation of algorithmic copyright enforcement on YouTube. We use digital and computational methods to help understand the operation of automated decision-making at scale. We argue that in order to understand complex, automated systems, we require new methods and research infrastructure to understand their operation at scale, over time, and across platforms and jurisdictions. We use YouTube takedowns as a case study to develop and test an innovative methodology for evaluating automated decision-making. First, we built technical infrastructure to obtain a random sample of 59 million YouTube videos and tested their availability two weeks after they were first published. We then used topic modeling to identify categories of videos for further analysis, and trained a machine learning classifier to categorise videos across the entire dataset. We then use statistical analysis (multinomial logistic regression) to examine the characteristics of videos that are most likely to be removed through DMCA notices, Content ID removals, and Terms of Service enforcement. This interdisciplinary work provides the methodological base for further experimentation with the use of deep neural nets to enable large-scale analysis of the operation of automated systems in the realm of digital media. We hope that this work will improve understanding of a useful and fruitful set of methods to interrogate pressing public policy research questions in the context of content moderation and automated decision-making.

scholarly journals Addressing Disconnection: Automated Decision-Making, Administrative Law and Regulatory Reform

2021 ◽  
Vol 44 (3) ◽  
Author(s):  
Anna Huggins
Keyword(s):  
Decision Making ◽  
Judicial Review ◽  
Regulatory Reform ◽  
Administrative Law ◽  
Government Agencies ◽  
Mental Process ◽  
Automated Systems ◽  
Regulatory Evolution ◽  
Distinctive Features ◽  
Automated Decision Making

Automation is transforming how government agencies make decisions. This article analyses three distinctive features of automated decision-making that are difficult to reconcile with key doctrines of administrative law developed for a human-centric decision-making context. First, the complex, multi-faceted decision-making requirements arising from statutory interpretation and administrative law principles raise questions about the feasibility of designing automated systems to cohere with these expectations. Secondly, whilst the courts have emphasised a human mental process as a criterion of a valid decision, many automated decisions are made with limited or no human input. Thirdly, the new types of bias associated with opaque automated decision-making are not easily accommodated by the bias rule, or other relevant grounds of judicial review. This article, therefore, argues that doctrinal and regulatory evolution are both needed to address these disconnections and maintain the accountability and contestability of administrative decisions in the digital age.

scholarly journals Algorithmic Accountability in Context. Socio-Technical Perspectives on Structural Causal Models

2021 ◽  
Vol 3 ◽  
Author(s):  
Nikolaus Poechhacker ◽  
Severin Kacianka
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Social Theory ◽  
Social System ◽  
Causal Models ◽  
The Social ◽  
Algorithmic Accountability ◽  
Research Questions ◽  
New Form ◽  
Automated Decision Making

The increasing use of automated decision making (ADM) and machine learning sparked an ongoing discussion about algorithmic accountability. Within computer science, a new form of producing accountability has been discussed recently: causality as an expression of algorithmic accountability, formalized using structural causal models (SCMs). However, causality itself is a concept that needs further exploration. Therefore, in this contribution we confront ideas of SCMs with insights from social theory, more explicitly pragmatism, and argue that formal expressions of causality must always be seen in the context of the social system in which they are applied. This results in the formulation of further research questions and directions.

Estimating shopping center visitor numbers based on a new hybrid fuzzy prediction method

2021 ◽  
pp. 1-14
Author(s):  
Cagatay Ozdemir ◽  
Sezi Cevik Onar ◽  
Selami Bagriyanik ◽  
Cengiz Kahraman ◽  
Burak Zafer Akalin ◽  
...  
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Large Scale ◽  
Prediction Method ◽  
Base Stations ◽  
Shopping Mall ◽  
Multi Criteria Decision Making ◽  
Shopping Malls ◽  
Data Production ◽  
Telecommunications Services

Companies started to determine their strategies based on intelligent data analysis due to stagey enhance data production. Literature reviews show that the number of resources where demand estimation, location analysis, and decision-making technique applied together with the machine learning method is low in all sectors and almost none in the shopping mall domain. Within this study’s scope, a new hybrid fuzzy prediction method has been developed that will estimate the customer numbers for shopping malls. This new methodology is applied to predict the number of visitors of three shopping malls on the Anatolian side of Istanbul. The forecasting study for corresponding shopping malls is made by using the daily signaling data from indoor base stations of large-scale technology and telecommunications services provider and the features to be used in machine learning models is determined by fuzzy multi criteria decision making method. Output revealed by the application of the fuzzy multi criteria decision making method enables the prioritization of features.

Using large-scale experiments and machine learning to discover theories of human decision-making

Science ◽  
2021 ◽  
Vol 372 (6547) ◽  
pp. 1209-1214
Author(s):  
Joshua C. Peterson ◽  
David D. Bourgin ◽  
Mayank Agrawal ◽  
Daniel Reichman ◽  
Thomas L. Griffiths
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Large Scale ◽  
Risky Choice ◽  
Machine Learning Algorithms ◽  
Psychological Theories ◽  
The Social ◽  
Power Machine ◽  
Human Decision ◽  
Gradient Based

Predicting and understanding how people make decisions has been a long-standing goal in many fields, with quantitative models of human decision-making informing research in both the social sciences and engineering. We show how progress toward this goal can be accelerated by using large datasets to power machine-learning algorithms that are constrained to produce interpretable psychological theories. Conducting the largest experiment on risky choice to date and analyzing the results using gradient-based optimization of differentiable decision theories implemented through artificial neural networks, we were able to recapitulate historical discoveries, establish that there is room to improve on existing theories, and discover a new, more accurate model of human decision-making in a form that preserves the insights from centuries of research.

Automated Decision-Making and Administrative Law

2020 ◽  
pp. 655-676
Author(s):  
Michèle Finck
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Deep Learning ◽  
Data Analysis ◽  
Computational Intelligence ◽  
Sustainable Use ◽  
Administrative Law ◽  
Guiding Principles ◽  
Automated Decision Making

This chapter examines the uses of automated decision-making (ADM) systems in administrative settings. First, it introduces the current enthusiasm surrounding computational intelligence before a cursory overview of machine learning and deep learning is provided. The chapter thereafter examines the potential of these forms of data analysis in administrative processes. In addition, this chapter underlines that, depending on how they are used; these tools risk impacting pejoratively on established concepts of administrative law. This is illustrated through the example of the principle of transparency. To conclude, a number of guiding principles designed to ensure the sustainable use of these tools are outlined and topics for further research are suggested.

scholarly journals Playing with machines: Using machine learning to understand automated copyright enforcement at scale

2020 ◽  
Vol 7 (1) ◽  
pp. 205395172091996
Author(s):  
Joanne E Gray ◽  
Nicolas P Suzor
Keyword(s):  
Machine Learning ◽  
Computational Methods ◽  
Large Scale ◽  
Detection System ◽  
Language Model ◽  
Levels Of Abstraction ◽  
Systematic Analysis ◽  
Learning Classifier ◽  
Large Scale Analysis ◽  
Copyright Enforcement

This article presents the results of methodological experimentation that utilises machine learning to investigate automated copyright enforcement on YouTube. Using a dataset of 76.7 million YouTube videos, we explore how digital and computational methods can be leveraged to better understand content moderation and copyright enforcement at a large scale.We used the BERT language model to train a machine learning classifier to identify videos in categories that reflect ongoing controversies in copyright takedowns. We use this to explore, in a granular way, how copyright is enforced on YouTube, using both statistical methods and qualitative analysis of our categorised dataset. We provide a large-scale systematic analysis of removals rates from Content ID’s automated detection system and the largely automated, text search based, Digital Millennium Copyright Act notice and takedown system. These are complex systems that are often difficult to analyse, and YouTube only makes available data at high levels of abstraction. Our analysis provides a comparison of different types of automation in content moderation, and we show how these different systems play out across different categories of content. We hope that this work provides a methodological base for continued experimentation with the use of digital and computational methods to enable large-scale analysis of the operation of automated systems.

scholarly journals Encoding Equality: The Case for Greater Regulation of Artificial Intelligence and Automated Decision Making in New Zealand

2020 ◽  
Vol 51 (1) ◽  
pp. 1
Author(s):  
Ella Brownlie
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Decision Making ◽  
Current Approach ◽  
Learning Processes ◽  
The Novel ◽  
General Data Protection Regulation ◽  
Urgent Case ◽  
General Data ◽  
Automated Decision Making

Automated decision-making systems, developed using artificial intelligence and machine learning processes, are being used by companies, organisations and governments with increasing frequency. The purpose of this article is to outline the urgent case for regulating automated decision-making and examine the possible options for regulation. This article will argue that New Zealand's current approach to regulating decision-making is inadequate. It will then analyse art 22 of the European Union's General Data Protection Regulation, concluding that this regime also has significant flaws. Finally, this article will propose an alternative regulatory solution to address the novel challenge posed by automated decision-making. This solution aims to strike a balance between the interests of organisations in capitalising on the benefits of automated decision-making technology and the interests of individuals in ensuring that their right to freedom from discrimination is upheld.

Sign in / Sign up

Export Citation Format

Share Document