Alzheimer's disease (AD) is known to be a multifactorial neurodegenerative disorder, and is one of the main causes of dementia in the elderly. Many studies have demonstrated molecules involved in the pathogenesis of AD, however its underlying mechanisms remain obscure. It may be simplistic to try to explain the disease based on the role of a few genes only. Accumulating new, huge amount of information from e.g. genome, proteome and interactome datasets and new knowledge, we are now able to clarify and characterize diseases essentially as a result of dysfunction of molecular networks. Recent studies have indicated that relevant genes affected in human diseases concentrate in a part of the network, often called as "disease module." In the case of AD, some disease-associated pathways seem different, but some of them are clearly disease-related and coherent. This suggests the existence of a common pathway that negatively drives from healthy state to disease state (i.e., the disease module(s)). Additionally, such disease modules should dynamically change through AD progression. Thus, network-level approaches are indispensable to address unknown mechanisms of AD. In this chapter, we introduce network strategies using gene co-expression and protein interaction networks.