Machine Decisions and Human Consequences

This chapter discusses how machine learning works and, as a result, how decisions are made by modern intelligent algorithms or ‘classifiers’. It critically evaluates a real-world ‘classifier’, the Harm Assessment Risk Tool (HART)—an algorithmic decision-making tool employed by the Durham police force to inform custody decisions concerning individuals who have been arrested for suspected criminal offences. It evaluates the tool by reference to four normative benchmarks: prediction accuracy, fairness and equality before the law, transparency and accountability, and informational privacy and freedom of expression. It argues that systems which utilize decision-making (or decision-supporting) algorithms, and have the potential to detrimentally affect individual or collective human rights, deserve significantly greater regulatory scrutiny than those systems that use decision-making algorithms to process objects.

Download Full-text

Sa1470 RISK ESTIMATION, MACHINE LEARNING BASED ERCP DECISION-MAKING TOOL FOR SUSPECTED CHOLEDOCHOLITHIASIS

Gastrointestinal Endoscopy ◽

10.1016/j.gie.2019.03.264 ◽

2019 ◽

Vol 89 (6) ◽

pp. AB246-AB247

Author(s):

Venkata S. Akshintala ◽

Bohao Tang ◽

Ayesha Kamal ◽

James L. Buxbaum ◽

B. Joseph Elmunzer ◽

...

Keyword(s):

Machine Learning ◽

Decision Making ◽

Risk Estimation ◽

Decision Making Tool

Download Full-text

Part 5 Cross-Cutting Issues, 5.2 Limitations

10.1093/law/9780198703983.003.0029 ◽

2016 ◽

Author(s):

Bielefeldt Heiner, Prof ◽

Ghanea Nazila, Dr ◽

Wiener Michael, Dr

Keyword(s):

Decision Making ◽

Human Rights ◽

National Security ◽

Freedom Of Expression ◽

Fair Trial ◽

Freedom Of Religion ◽

Freedom Of Association

This chapter discusses issues concerning the limitation of freedom of religion or belief, including related issues of interpretation. First, religious manifestation is different from holding, adopting, or changing religion or belief in that the latter is not subject to any limitation whatsoever. Second, national security is mentioned as a limitations ground in articles 12 (liberty of movement), 13 (expulsion), 14 (fair trial), 19 (freedom of expression), 21 (peaceful assembly), and 22 (freedom of association), but not in article 18 of the ICCPR. Third, traditions and limitations intersect in two ways; a single tradition cannot determine limitation on the grounds of morals, and limitations should not target a single tradition. Lastly, the exercise of assessing ‘necessity’ and ‘proportionality’ should not sideline the importance of upholding the protection of the relevant human rights. International procedures should only limit freedom of religion or belief in accordance with a strict understanding of the rights and limitation regime concerned, and with detailed justification of the rationale for their decision making in the compromise that is proposed.

Download Full-text

The Politics of Hate Speech: A Case Comment on Warman v Lemire

Constitutional Forum / Forum constitutionnel ◽

10.21991/c9sx0d ◽

2012 ◽

Vol 19 (1, 2 & 3) ◽

pp. 2011

Author(s):

Ranjan K. Agarwal

Keyword(s):

Decision Making ◽

Human Rights ◽

Supreme Court ◽

Hate Speech ◽

Freedom Of Expression ◽

Court Ruling ◽

Supreme Court Ruling ◽

Teleological Reasoning ◽

Academic Debate

In September 2009, the Canadian Human Rights Tribunal waded into a highly public and acrimonious debate about the role of human rights tribunals and commissions, especially in policing hate speech. In Warman v Lemire,1 the Tribunal held that section 13(1) of the Canadian Human Rights Act2 (CHRA), which prohibits the communication of hate messages, infringed the constitutional guarantee of freedom of expression, section 2(b) of the Charter of Rights and Freedoms.3 The decision added to a firestorm of media, political and academic debate about whether anti-discrimination statutes should prohibit hate speech. The Warman decision is complicated by a twenty-year-old Supreme Court ruling, in a 4–3 decision, that a predecessor provision in the CHRA is constitutional. In this article, I argue that the Tribunal’s decision is logically unsound and likely the result of ends-based or teleological reasoning. In my view, ends-based reasoning does not assist in Charter analysis as it produces decisions that call into question the legitimacy of the courts. This article first outlines the facts in Warman and the Tribunal’s holding on the constitutional issues. It goes on to survey the legal and constitutional background to the Warman decision and discuss the Taylor precedent. It then describes the Tribunal’s reasoning on constitutional issues, including the Taylor decision and amendments to the CHRA after Taylor. Finally, it criticizes the Tribunal’s ends-based reasoning and argues that this type of reasoning is illegitimate in constitutional decision-making.

Download Full-text

Decision Making by a Multiple-Rule Classifier

Encyclopedia of Decision Making and Decision Support Technologies ◽

10.4018/978-1-59904-843-7.ch020 ◽

2011 ◽

pp. 176-183

Author(s):

Ivan Bruha

Keyword(s):

Machine Learning ◽

Decision Making ◽

Knowledge Discovery ◽

Learning Algorithm ◽

Decision Rules ◽

Mode Decision ◽

Machine Learning Algorithm ◽

Statistical Tools ◽

Conflict Situations ◽

Decision Supporting

A rule-inducing learning algorithm yields a set of decision rules that depict knowledge discovered from a (usually large) dataset; therefore, this topic is often known as knowledge discovery from databases (KDD). Any classifier (or, expect system) then can utilize this decision set to derive a decision about given problems, observations, or diagnostics. The decision set (induced by a learning algorithm) may be either of the form of an ordered or unordered set of rules. The latter seems to be more understandable by humans and directly applicable in most expert systems, or generally, any decision- supporting one. However, classification utilizing the unordered-mode decision set may be accompanied by some conflict situations, particularly when several rules belonging to different classes match (are satisfied by, “fire” for) an input to-be-classified (unseen) object. One of the possible solutions to this conflict is to associate each decision rule induced by a learning algorithm with a numerical factor, which is commonly called the rule quality (An & Cercone, 2001; Bergadano et al., 1988; Bruha, 1997; Kononenko, 1992; Mingers, 1989; Tkadlec & Bruha, 2003). This article first briefly introduces the underlying principles for defining rules qualities, including statistical tools such as contingency tables and then surveys empirical and statistical formulas of the rule quality and compares their characteristics. Afterwards, it presents an application of a machine learning algorithm utilizing various formulas of the rule qualities in medical area.

Download Full-text

A Machine Learning decision-making tool for extubation in Intensive Care Unit patients

Computer Methods and Programs in Biomedicine ◽

10.1016/j.cmpb.2020.105869 ◽

2020 ◽

pp. 105869

Author(s):

Alexandre Fabregat ◽

Mónica Magret ◽

Josep Anton Ferré ◽

Anton Vernet ◽

Neus Guasch ◽

...

Keyword(s):

Machine Learning ◽

Intensive Care Unit ◽

Decision Making ◽

Intensive Care ◽

Decision Making Tool

Download Full-text

Mo1176 DEVELOPMENT OF A MACHINE LEARNING (ML) BASED DECISION MAKING TOOL TO DETERMINE RISK AND TIMING OF RECURRENCE OF BARRETT'S ESOPHAGUS (BE) AFTER COMPLETE ERADICATION OF INTESTINAL METAPLASIA (CE-IM)

Gastroenterology ◽

10.1016/s0016-5085(20)32717-7 ◽

2020 ◽

Vol 158 (6) ◽

pp. S-814-S-815

Author(s):

Venkata S. Akshintala ◽

Haley Grant ◽

Samuel Han ◽

Albert Kuo ◽

Vladimir Kushnir ◽

...

Keyword(s):

Machine Learning ◽

Decision Making ◽

Intestinal Metaplasia ◽

Barrett’S Esophagus ◽

Barrett's Esophagus ◽

Complete Eradication ◽

Decision Making Tool

Download Full-text

593 – Development, Validation of a Post-Ercp Pancreatitis Risk Calculator and Machine Learning Based Decision Making Tool for Prophylaxis Selection

Gastroenterology ◽

10.1016/s0016-5085(19)37084-2 ◽

2019 ◽

Vol 156 (6) ◽

pp. S-116

Author(s):

Venkata S. Akshintala ◽

Albert Kuo ◽

Ayesha Kamal ◽

B. Joseph Elmunzer ◽

Hui Luo ◽

...

Keyword(s):

Machine Learning ◽

Decision Making ◽

Risk Calculator ◽

Decision Making Tool

Download Full-text

The electoral commission, disinformation and freedom of expression

Northern Ireland Legal Quarterly ◽

10.53386/nilq.v71i4.543 ◽

2021 ◽

Vol 71 (4) ◽

pp. 675-681

Author(s):

Ethan Shattock

Keyword(s):

Human Rights ◽

Potential Role ◽

Freedom Of Expression ◽

European Convention ◽

European Court ◽

Starting Point ◽

Regulatory Scrutiny ◽

Run Up ◽

The Right

This commentary examines how the prospective electoral commission could play a role in combatting disinformation in the run-up to Irish elections. While legislative debates have pointed to the potential role of the commission in protecting elections from anti-democratic actors who disseminate false electoral claims, no clear mandate has detailed how this could manifest. This ambiguity is exacerbated by Ireland’s electoral statutory framework, which has struggled to adapt to the challenging digital realities of contemporary electoral engagement. While the emergence of disinformation and related digital exigencies represents a potential for regulatory scrutiny, this must be considered alongside Article 10 of the European Convention on Human Rights (ECHR) and Article 40.6.1 of the Irish Constitution, both of which protect the right to freedom of expression. In positing how the new commission could counter electoral disinformation, a natural starting point is to probe how such functions are shaped and limited by this fundamental right. Moreover, the reluctance of the Irish judiciary and the European Court of Human Rights (ECtHR) to accept regulatory interference with political expression means that restrictions on the dissemination of information in the run-up to elections must be treated with delicacy when shaping the commission’s potential functions in this critical area.

Download Full-text