Knowledge on daily human activities in various domains is invaluable for many customized user services that can benefit from context-awareness or activity predictions. Past approaches to constructing a knowledge base of this kind have been domain-specific and not scalable. A recent attempt to extract activities of daily living (ADL) from Web resources deal with activities and objects involved in achieving them but not the sequence of actions in an activity. This paper describes an approach to automatically extracting human activity knowledge from Web articles that describe methods for performing tasks in a variety of domains. The target knowledge base is comprised of activity goals, actions, and ingredients, which are extracted with syntactic pattern-based and probabilistic machine learning based methods. The result is evaluated for accuracy and coverage against some baselines.