Approximate member query method based on high-dimensional data filter

The invention discloses an approximate member query method based on a high-dimensional data filter. The approximate member query method based on the high-dimensional data filter is capable of, through defining new distance sensitive hash function, respectively representing multi-dimensional data in a target data set and the to-be-queried multi-dimensional data without rebuilding the filter, supporting the approximate member query of more filtration distance parameters, greatly reducing the space cost, judging in the manner of 'and-or' combination while finally determining the approximate member of target data set omega through multiple function class numbers are used and each function class number contains multiple functions, and reducing the false negative rate of the filter. In addition, the approximate member query method based on the high-dimensional data filter is capable of, through building an entity checker and checking, greatly reducing the false positive rate of the filters with different filtration distance parameters.