liveBook · Manning

What you will learn in Hidden Influences

How algorithmic recommenders shape our lives

by Luca Belli · Manning Early Access Program (MEAP)

Chapter by chapter

Chapter 1: Recommender systems in action

The introduction illustrates how user behavior and recommender systems can reinforce each other and potentially steer people toward more extreme content.

1.1 Recommender systems are everywhere · 1.1.1 A world without RecSys · 1.1.2 What exactly are RecSys? · 1.2 Social media: From bulletin boards to doom scrolling · 1.2.1 The explosive success of social media · 1.2.2 RecSys in social media · 1.3 Societal impact · 1.3.1 The influence of platforms · 1.3.2 Recommendation as amplification · 1.3.3 Measuring recommender systems · 1.3.4 Algorithmic amplification as a measure of societal impact · 1.4 Summary

What you learn in chapter 1 →

Chapter 2: Anatomy of a recommender system

This introduction presents the black box model of a recommender system as a way to focus on its input and output.

2.1 The user-and-item world · 2.2 Parts of a RecSys · 2.2.1 Candidate generation · 2.2.2 Filtering · 2.2.3 Heavy ranking · 2.2.4 Post processing · 2.2.5 Delivering and recording · 2.3 Heavy ranking revisited: Training a RecSys · 2.3.1 Heuristics · 2.3.2 Data-driven approaches · 2.4 How to train a machine learning model · 2.5 Raw and derived scores · 2.6 Platforms’ design and amplification · 2.7 What happens when you refresh your feed · 2.8 Summary

What you learn in chapter 2 →

Chapter 3: Content moderation

This introduction defines content moderation and highlights the challenges of applying platform policies, including contextual interpretation and legal limits.

3.1 Why platforms moderate · 3.2 The impossibility of moderation at scale · 3.3 Content taxonomy · 3.3.1 Illegal content · 3.3.2 Content in violation of the terms of service · 3.3.3 Restricted content · 3.3.4 The rest of the content · 3.4 How content moderation is done · 3.4.1 The use of automation in content moderation · 3.4.2 Decisions and consequences · 3.4.3 Appeals · 3.5 Challenges in content moderation · 3.5.1 Lack of precedents · 3.5.2 Impact on moderators · 3.5.3 Cultural context · 3.5.4 Nuanced decisions · 3.5.5 Legal landscape · 3.5.6 Evolving field · 3.6 Lack of transparency · 3.7 Legal basis · 3.8 Summary

What you learn in chapter 3 →

Chapter 4: Algorithmic amplification: Measuring RecSys

This introduction explains how recommender systems distribute attention and why assessing their amplification requires difficult judgments about preferences, quality, and fairness.

4.1 Quantifying amplification · 4.2 Implicit learning · 4.3 Users’ preferences and proxy metrics · 4.4 Testing recommender systems: A/B tests · 4.5 Lack of a baseline · 4.6 Socio-technical nature of technology · 4.6.1 Feedback loops · 4.6.2 Feedback loops in recommender systems and social media · 4.6.3 The rich-getting-richer effect · 4.7 Labeling · 4.7.1 Labeling content · 4.7.2 Labeling users · 4.8 Summary

What you learn in chapter 4 →

Chapter 5: Redefining amplification

This introduction frames amplification measurement as a consequential question of power and a way to understand whether fringe views reach the mainstream.

5.1 The triangle and the elephant: Measuring amplification · 5.2 The importance of measuring amplification · 5.3 Who should define amplification? · 5.4 The journey of a piece of content · 5.5 Operationalizing amplification · 5.5.1 Measuring group amplification · 5.5.2 Producer side amplification · 5.5.3 Reader-side amplification: Feed analysis · 5.6 Summary

What you learn in chapter 5 →

Chapter 6: The business of RecSys: how platforms make money

This introduction outlines platform revenue models and the ways recommendations can support them.

6.1 Subscription · 6.2 Increased volume · 6.3 Profit cut · 6.4 Attention economy · 6.4.1 Ads and users’ surveillance · 6.4.2 Tracking across the internet · 6.4.3 Promoted content · 6.5 Mixture of models · 6.6 Summary

What you learn in chapter 6 →

Chapter 7: Societal impact

This introduction frames societal impacts as technology-attributable changes, including spillovers, and cautions against simplistic claims about recommender systems’ role.

7.1 Social contagion · 7.2 Boosting voter turnout · 7.3 The risk of information pollution: genocide · 7.4 Organizing online · 7.4.1 #BlackLivesMatter · 7.4.2 #MeToo · 7.4.3 Meme wars · 7.5 The effect on culture · 7.5.1 News · 7.5.2 Homogenization · 7.6 Summary

What you learn in chapter 7 →

Chapter 8: The unreasonable effectiveness of engagement

This introduction frames engagement-based recommender systems as socio-technical systems shaped by users, technology, and historical context.

8.1 The people · 8.1.1 Group belonging · 8.1.2 The search for novelty · 8.2 The technology · 8.2.1 Communities and like-minded users · 8.2.2 Out-of-Network content · 8.2.3 Feedback loops and extreme content · 8.2.4 Rabbit holing · 8.2.5 The firehose problem: scale and speed · 8.3 Society · 8.4 Summary

What you learn in chapter 8 →

Chapter 9: Understanding platform incentives

This introduction connects financial and growth incentives to technology companies’ practices and their tendency to overlook social risks.

9.1 The pressure from above: VC funding · 9.1.1 Creating and scaling virtual prototypes · 9.1.2 Growing companies and their evaluations · 9.1.3 Enshittification · 9.2 Silicon Valley’s role in defining the playbook · 9.2.1 Perpetuating blind spots: a tight-knit community · 9.2.2 Making products people want: the design principles · 9.2.3 Finding yet another problem to solve · 9.2.4 The obsession with simple metrics · 9.2.5 The great lie: making the World a better place · 9.3 How the alarm bells got ignored · 9.4 Summary

What you learn in chapter 9 →

Chapter 10: Mitigating impacts: platforms, individuals, and society

The introduction recaps recommender systems’ benefits and risks and argues that mitigating platform harms requires collective as well as individual action.

10.1 Platforms: power and duty over users · 10.1.1 Power over users, and lack of contestability · 10.1.2 The power of incentives · 10.1.3 Duty of care · 10.2 Taking action as individuals · 10.2.1 Seeing through the lie: meta awareness · 10.2.2 Disrupt the disruption · 10.2.3 Reclaim privacy · 10.2.4 “On the internet, nobody knows you are a dog” · 10.3 E pluribus unum: collective action to ensure customer protection · 10.3.1 Platform portability · 10.3.2 Data sharing · 10.4 Summary

What you learn in chapter 10 →

Concepts explained in this book

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.