Real-world physical objects and abstract data entities are interconnected, forming gigantic networks.By structuring these objects and their interactions into multiple types, such networks becomesemi-structured heterogeneous information networks. Most real-world applications that handle bigdata, including interconnected social media and social networks, scientific, engineering, or medicalinformation systems, online e-commerce systems, and most database systems, can be structuredinto heterogeneous information networks. Therefore, effective analysis of large-scale heterogeneousinformation networks poses an interesting but critical challenge.In my thesis, I investigate the principles and methodologies of mining heterogeneous informationnetworks. Departing from many existing network models that view interconnected data ashomogeneous graphs or networks, our semi-structured heterogeneous information network modelleverages the rich semantics of typed nodes and links in a network and uncovers surprisingly richknowledge from the network. This semi-structured heterogeneous network modeling leads to aseries of new principles and powerful methodologies for mining interconnected data, including (1)ranking-based clustering, (2) meta-path-based similarity search and mining, (3) user-guided relationstrength-aware mining, and many other potential developments. This thesis introduces thisnew research frontier and points out some promising research directions.