This chapter introduces machine learning as a subset of artificial intelligence that enables computers to learn from data without explicit programming. It defines machine learning using Tom Mitchell’s formal framework and explores practical applications like self-driving cars, optical character recognition, and recommendation systems. The chapter focuses on regression as a fundamental machine learning technique, explaining linear regression for modeling relationships between variables. A key section covers gradient descent, an optimization algorithm that iteratively finds the best model parameters by minimizing error functions. Through hands-on Python examples, students learn to implement both linear regression and gradient descent algorithms, visualizing how models improve over iterations. The chapter emphasizes practical considerations for choosing appropriate algorithms, including accuracy, training time, linearity assumptions, and the number of parameters, preparing students for more advanced supervised and unsupervised learning techniques.

This chapter focuses on data collection methods, analysis approaches, and evaluation techniques in data science. It covers various data collection methods including surveys (with different question types like multiple-choice, Likert scales, and open-ended questions), interviews, focus groups, diary studies, and user studies in lab and field settings.

The chapter distinguishes between quantitative methods (using numerical measurements and statistical analysis) and qualitative methods (observing behaviors, attitudes, and opinions through techniques like grounded theory and constant comparison). It also discusses mixed-method approaches that combine both methodologies.

For evaluation, the chapter explains model comparison metrics including precision, recall, F-measure, ROC curves, AIC, and BIC. It covers validation techniques like training-testing splits, A/B testing, and cross-validation methods. The chapter emphasizes that data science involves pre-data collection planning and post-analysis evaluation, not just data processing.

This chapter analyses efforts within the United Nations to develop legal and normative frameworks for transnational corporations (TNCs) and human rights, beginning in the 1970s. It first considers the UN Code of Conduct for Transnational Corporations and explains why this initiative failed to materialise despite many years of negotiation. It then examines the Global Compact, which reflects emerging trends in legalisation through its emphasis on implementation, participation by non-state actors, and reliance on consensus-building and norm promotion. The chapter next reviews the rise and fall of the Draft Norms, before turning to the development of the UN Guiding Principles on Business and Human Rights. This section highlights the innovative nature of Ruggie’s constructivist approach to generating new legal and social norms. A new treaty process, initiated in 2014, remains ongoing and suggests that traditional legalisation strategies continue to retain relevance in certain contexts.

This chapter explores the evolution and techniques of corporate tax arbitrage, focusing on how multinational corporations (MNCs) exploit the structural features of the international tax regime. It traces the origins of the system to the 1928 League of Nations framework, which privileged source and residence taxation while neglecting valuation and the treatment of intangible assets. This omission, combined with the legal autonomy granted to subsidiaries within centrally coordinated multi-corporate enterprises, created enduring arbitrage opportunities. The chapter analyses how firms leverage intangible assets, ownership structures, and corporate residency rules to reallocate profits across jurisdictions, exploiting regulatory mismatches. It highlights the role of offshore financial centres (OFCs), intermediary subsidiaries, and hybrid instruments in enabling tax avoidance. Case studies – such as Apple’s use of stateless entities and Amazon’s transfer of losses via Luxembourg – illustrate how MNCs circumvent tax rules through entity design, subsidiary chaining, and strategic use of global value chains. Using new evidence from the CORPLINK study, the chapter estimates that although OFC intermediaries represent only 1.7 per cent of subsidiaries among the world’s top 100 non-financial firms, they control up to two-thirds of group revenues. Tax arbitrage is shown to be systemic, not exceptional, and embedded in contemporary corporate structures.

This chapter introduces Python as a powerful yet beginner-friendly programming language essential for data science. It covers getting access to Python through direct installation or integrated development environments like Anaconda and Spyder. The chapter teaches fundamental programming concepts including basic operations, data types, and key data structures (lists, tuples, dictionaries, sets, and DataFrames). Students learn to write control structures using if-else statements and while/for loops, create reusable functions, and make programs interactive through user input. The chapter also explains how to install and use Python packages, which extend the language’s capabilities for specialized tasks. Throughout, practical examples demonstrate concepts like leap year calculations, temperature categorization, and sales data analysis. The chapter emphasizes Python’s accessibility, extensive package ecosystem, and suitability for data science applications, positioning it as an ideal tool for solving computational and data analysis problems.

This chapter covers unsupervised learning, where algorithms analyze data without known true labels or outcomes. Unlike supervised learning, the goal is to discover hidden patterns and structures in data.

The chapter explores three main techniques: Agglomerative clustering works bottom-up, starting with individual data points and merging similar ones into larger clusters. Divisive clustering (including k-means) takes a top-down approach, splitting data into smaller groups. Both methods use distance matrices and dendrograms to visualize cluster relationships.

Expectation Maximization (EM) handles incomplete data by iteratively estimating missing parameters using maximum likelihood estimation. Model quality is assessed using AIC and BIC criteria.

The chapter also introduces reinforcement learning, where agents learn optimal actions through trial-and-error interactions with environments, receiving rewards or penalties. Applications include robotics, gaming, and autonomous systems. Throughout, the chapter emphasizes the creative, interpretive nature of unsupervised learning compared to more structured supervised approaches.

The book examines the various arenas in which actors are making – and breaking – the rules in business and human rights. It advances a framework for analysing these developments by adapting the liberal institutionalist concept of legalisation articulated in Kenneth Abbott et al.’s article ‘The Concept of Legalization’. Applied in the transnational context, the classic framework appears incomplete: it omits a crucial dimension – implementation – which operates alongside obligation, precision and delegation. The empirical chapters in this book reveal that efforts toward implementation are often pursued with the aim of strengthening one or more of the other dimensions over time. In such cases, actors play the long game: they may accept lower levels of obligation, precision or delegation in the short term, anticipating that early attention to implementation will enhance these dimensions in the longer run. Beyond business and human rights, this revised framework may also illuminate regulatory dynamics in transnational fields such as climate governance, national security, and anti-trafficking.

In the framework of the common objective of this volume, this chapter focuses on the technological element –expressed in AI– which is usually part of the definition of remote work. This chapter discusses how AI tools shape the organization and performance of remote work, how algorithms impact remote workers rights and how trade unions and workers can harness these powerful instruments to improve working and living conditions. Three hypotheses are considered. First, that AI systems and algorithmic management generate a de facto deepening of the subordinate position of the worker. Second, that this process does not represent technological determinism but instead the impact of human and institutional elements. And finally, that technological resources usually are more present in remote work than in traditional work done at the workplace. These hypotheses and concerns are addressed in several ways: by contextualizing the issue over time, through a multi-level optic centered on the interactions of different levels of regulation, by examining practical dimensions and finally by exploring the implications for unions and worker agency.

This chapter argues that fundamental problems limit ESG’s potential benefits for society and can be traced back to ESG’s initial conceptualization in the early 2000s in the advent of the United Nation’s Global Compact initiative. ESG from the very beginning has been built, on the one hand, on the premise of promoting institutional investors’ interests at the expense of critical stakeholders’ concerns and, on the other hand, on quite idealistic assumptions about the proper functioning of markets and states. Drawing from the theory of deliberative democracy, this chapter develops suggestions of how ESG could become more beneficial to people and planet by making the ESG investing system, understood as an organized set of actors and procedures, more inclusive, argumentative, and consequential with a view on societal rather than investors’ benefits. The chapter proposes that incorporating deliberation in the governance structure of rating agencies specifically is one way to do so.

This preface introduces “A Hands-On Introduction to Data Science with Python,” designed for advanced undergraduates and graduate students across diverse fields including information science, business, psychology, and sociology. The book requires minimal programming experience but expects computational thinking and basic statistics knowledge. It’s structured in four parts: foundations of data science, tools and platforms (Python programming and cloud computing), machine learning techniques, and real-world applications. The second edition adds "DS in Practice" boxes, cloud computing coverage, AI/ethics discussions, and downloadable datasets. The book emphasizes practical, hands-on learning with 39 solved exercises, 40 try-it-yourself problems, and 57 end-of-chapter problems, making data science accessible to non-technical students.

The incorporation of environmental, social, and governance (ESG) criteria into corporate strategies has become a prominent feature of the modern business landscape. As part of this movement, there is an increasing trend toward the monetary valuation of sustainability impacts. While the intention behind assigning monetary values to ESG-related impacts may be to provide a quantifiable basis for decision-making, it raises profound ethical concerns. This chapter explores the ethical dilemmas surrounding the monetary valuation of sustainability impacts, especially within the broader context of ESG performance measurement, in three problem dimensions: (1) the commodification of life and nature, (2) unequal power dynamics and neocolonial features of ESG and valuation, and (3) the marginalization of unquantifiable impacts and intrinsic values. The chapter ends by exploring the moral hazards that come with ignoring these ethical problems, and how corporate responsibility and accountability mechanisms can take these into account.