Traditional Chinese Medicine (TCM) is characterized by its holistic thinking in treating a patient as a holistic system, and the use of various herbal formulae in restoring the balance of human body. The holistic philosophy and patient-tailored treatment of TCM shares a lot with the key idea of current systems biology and precision medicine. However, it is very challenging to unveil the biological basis of TCM by current reductionist "one gene, one target, one drug" approach due to the complexity of both the chemical compositions of herb formula and the biological systems of patients. To understand TCM from a systems perspective, we proposed a new concept and established a new approach of “network target, multi-component drug”, which emphasizes that treating the biomolecular networks underlying diseases or TCM Syndrome as a therapeutic target to determine the systematic intervention such as herbal formula. Then, by using artificial intelligence methods, we created a set of algorithms to predict the network target underlying disease and herbal formula. The effectiveness of the network target approach has been demonstrated by the follow-up clinical and experimental investigations. Typically, we constructed for the first time the metabolism-immune imbalanced molecular network underlying Cold / Hot syndrome, a pair of thousand-year-old concepts in TCM, and validated the network in a cohort of patients with gastritis and precancerous lesions. Furthermore, we identified the novel actions of compounds, network regulation mechanisms, and modern indications of Cold / Hot herbs in fighting inflammation and cancer from a classic “Nourishing Yin” formula, Liu-wei-di-huang, a “Clearing away Hot” formula, Qing-Luo-Yin, and various herbs. Taken together, the “network target”-based network pharmacology approach is expected to offer bright prospects in understanding TCM, narrow the gap between Eastern and Western medical practices, and help achieve the TCM-based systems and precision medicine in such a Big Data and Artificial Intelligence era.