The hash method is a fast algorithm for approximating near-neighbor search. It can effectively reduce the storage space and computation time. It is an important method for large data processing. It can be used in the fields of large-scale machine learning, image annotation and image retrieval. This report will provide a brief introduction to some of our research on the hash method, including data-independent methods and data-related methods, localized sensitive hash methods involving data types such as vectors, collections, subspaces, and learning-based hash method.