The first comprehensive source on cost-sensitive learning, this reference focuses on various research domains in machine learning that attempt to solve cost-sensitive learning with simplifying assumptions. The book explicitly deals with tradeoffs between the various costs and benefits inherent in many domains of machine learning research, including semisupervised learning, active learning, active feature acquisition, and inductive transfer learning. It covers recently developed methods, along with several applications of cost-sensitive learning, including natural language processing, computer vision, medical applications, web ads, and marketing. Real-world applications aid with reader understanding.