Publications
Welcome to the publication repository on Behavioural Data Science. Here you will find scientific as well as popular press articles, which will help you to better understand the field. This month, our highlights include:
Big Bad Bias Book
Looking to make better decisions and avoid common cognitive pitfalls? Look no further than this comprehensive guide! With over 200 biases of human decision-making systematized and analyzed, this book is the ultimate resource for anyone seeking to understand the intricacies of the human mind. Each bias is explored in detail, complete with fun trivia and real-world business use cases to help you apply the insights to your own life. Plus, you'll learn about the latest research on each bias and just how robust and replicable it truly is. And if you're looking to dive deeper, each chapter provides a list of additional literature for further study. Don't let biases hold you back - get this book and start making more informed decisions today!
The Rural Areas Missing out on AI Opportunities
This article discusses the potential discrepancies between urban and rural communities in the study and application of artificial intelligence (AI). Data collection from smartphones, which is predominantly harvested from rural populations, results in technologies being created to enrich urban lives. As such, rural communities are missing out on the benefits of data-driven research. AI and Behavioural Data Science has the potential to improve country life, including disaster management. Poor digital infrastructure and difficulty recruiting AI talent to rural areas are the main reasons for the gap. Local communities, researchers, and government must work together to ensure inclusive AI, benefitting rural and regional communities.
Improving Productivity in Hollywood with Data Science: Using Emotional Arcs of Movies to Drive Product and Service Innovation in Entertainment Industries
This paper proposes a framework for using behavioral data science to optimize content generation in the entertainment industry. The proposed approach uses natural language processing combined with econometric analysis to explore how emotions shape consumer preferences for media content, and how this affects revenue streams. The emotional trajectory of each motion picture is analyzed and grouped into clusters to predict the overall success parameters of movies, including box office revenues, viewer satisfaction, awards, and reviews. The study finds that emotional arcs in movies can be partitioned into six basic shapes, with the "Man in a Hole" shape associated with the highest box office success. Implications for improving productivity in the entertainment industry are discussed.
A COVID-19-Based Modified Epidemiological Model and Technological Approaches to Help Vulnerable Individuals Emerge from the Lockdown
This paper proposes a new approach to pandemic prevention using behavioural data science segmentation. The aim is to understand the variance in severity of COVID-19 across different population groups and model the risk levels of each group to enable targeted prevention strategies. The proposed SEIR-v model separates the population into two groups based on vulnerability to the disease and simulates the spread of the epidemic with different containment measures. Results suggest that a decrease in exposure of vulnerable individuals could significantly reduce the number of deaths caused by the disease. The study proposes the use of wristbands for vulnerable individuals and combining contact tracing data from smartphone apps to protect them.