Critics from across the political spectrum attack social media platforms for invading personal privacy. Social media firms famously suck in huge amounts of information about individuals who use their services (and sometimes others as well), and then monetize this data, primarily by selling targeted advertising. Many privacy advocates object to the very collection and use of this personal data by platforms, even if not shared with third parties. In addition, there is the ongoing (and reasonable) concern that the very existence of Big Data creates a risk of leaks. Further, aside from the problem of Big Data, the very existence of social media enables private individuals to invade the privacy of others by widely disseminating personal information. That social media firms’ business practices compromise privacy cannot be seriously doubted. But it is also true that Big Data lies at the heart of social media firms’ business models, permitting them to provide users with free services in exchange for data which they can monetize via targeted advertising. So unless regulators want to take free services away, they must tread cautiously in regulating privacy.

The area where social media has undoubtedly been most actively regulated is in their data and privacy practices. While no serious critic has proposed a flat ban on data collection and use (since that would destroy the algorithms that drive social media), a number of important jurisdictions including the European Union and California have imposed important restrictions on how websites (including social media) collect, process, and disclose data. Some privacy regulations are clearly justified, but insofar as data privacy laws become so strict as to threaten advertising-driven business models, the result will be that social media (and search and many other basic internet features) will stop being free, to the detriment of most users. In addition, privacy laws (and related rules such as the “right to be forgotten”) by definition restrict the flow of information, and so burden free expression. Sometimes that burden is justified, but especially when applied to information about public figures, suppressing unfavorable information undermines democracy. The chapter concludes by arguing that one area where stricter regulation is needed is protecting children’s data.

Since the 1990s, big data has rapidly grown, influencing business, government, and healthcare. Fueled by networked devices, social media, and affordable cloud storage, it features voluminous datasets with diverse types, rapid updates, and accuracy concerns. Applications span retail, manufacturing, transportation, finance, and education, yielding benefits like data-driven decisions, optimization, personalized marketing, scientific progress, and fraud detection. However, challenges arise from complexity, necessitating interdisciplinary collaboration, privacy issues, potential cyberattacks, and the need for robust data protections. Accurate interpretation is crucial, given the risk of costly misinterpretations. Moreover, significant resources for storage, processing, and analysis raise environmental concerns, while legal and ethical considerations add complexity. Overreliance on data may lead to missed opportunities, underscoring the importance of balancing insights with human judgment. In conclusion, big data offers immense potential but poses significant challenges. Navigating this landscape requires a nuanced understanding, fostering responsible data practices to unlock its potential for informed decision-making and advancements across diverse fields.

This chapter addresses how one could quantify and explore the impact of geopolitics on global businesses. Computational geopolitics is an attempt to integrate quantitative methods and geopolitical analysis to understand and predict trends. The explosive growth of data, improvements in computational power, and access to cloud computing have led to a proliferation of computational methods in analyzing geopolitics and its impact on companies. The chapter explores some tools and techniques used in computational geopolitics, including events-based approaches to measuring geopolitical tensions, textual approaches, and empirical approaches. In addition, it provides examples of ways in which analysts can quantify the impact of geopolitics on trade and foreign direct investment. It also introduces experimental methods to assess the effectiveness of companies’ strategic responses to geopolitical tensions. Large language models (LLMs) can be used for sentiment analysis, spotting trends, scenario building, risk assessment, and strategic recommendations. While they methods offer advances in quantifying the impact of geopolitics on global businesses, analysts should also be cautious about data quality and availability as well as the complexity of the phenomenon and the geopolitics of AI. The chapter concludes by pointing the reader to some widely used data sources for computational geopolitics.

A distinction between types of methods (understanding and explanation) that generate different kinds of evidence relevant to the psychiatric assessment is characterised. The distinction is animated with both non-clinical and clinical examples and exercises. Scepticism about the distinction is addressed, and three influential systems of psychiatric knowledge which collapse understanding and explanation in different ways are discussed. The argument is made that the distinction (analogous to the romantic/classic distinction) resurfaces and is compelling. However, another challenge becomes important – holism in psychiatric assessment – which the understanding/explanation distinction leaves in an unsatisfactory state.

While previous chapters discussed deep learning recommender systems from a theoretical and algorithmic perspective, this chapter shifts focus to the engineering platform that supports their implementation. Recommender systems are divided into two key components: data and model. The data aspect involves the engineering of the data pipeline, while the model aspect is split between offline training and online serving. This chapter is structured into three parts: (1) the data pipeline framework and big data platform technologies; (2) popular platforms for offline training of recommendation models like Spark MLlib, TensorFlow, and PyTorch; and (3) online deployment and serving of deep learning recommendation models. Additionally, the chapter covers the trade-offs between engineering execution and theoretical considerations, offering insights into how algorithm engineers can balance these aspects in practice.

The digital age, characterized by the rapid development and ubiquitous nature of data analytics and machine learning algorithms, has ushered in new opportunities and challenges for businesses. As the digital evolution continues to reshape commerce, it has empowered firms with unparalleled access to in-depth consumer data, thereby enhancing the implementation of a variety of personalization strategies. These strategies utilize sophisticated machine learning algorithms capable of attaining personal preferences, which can better tailor products and services to individual consumers. Among these personalization strategies, the practice of personalized pricing, which hinges on leveraging customer-specific data, is coming to the forefront.

The criteria for evaluating research studies often include large sample size. It is assumed that studies with large sample sizes are more meaningful than those that include a fewer number of participants. This chapter explores biases associated with the traditional application of null hypothesis testing. Statisticians now challenge the idea that retention of the null hypothesis signifies that a treatment is not effective. A finding associated with an exact probability value of p = 0.049 is not meaningfully different from one in which p = 0.051. Yet the interpretation of these two studies can be dramatically different, including the likelihood of publication. Large studies are not necessarily more accurate or less biased. In fact, biases in sampling strategy are amplified in studies with large sample sizes. These problems are of increasing concern in the era of big data and the analysis of electronic health records. Studies that are overpowered (because of very large sample sizes) are capable of identifying statistically significant differences that are of no clinical importance.