The invention provides a mining apparatus of activity patterns of individuals. The apparatus comprises a computing unit, a dividing unit and a clustering unit. The computing unit is configured to compute points of stay of an individual according to historical positional data of the individual, thus forming a point-of-stay sequence containing time of stay. The dividing unit is configured to divide the point-of-stay sequence into corresponding point-of-stay sequence segments according to an activity theme of the individual. The clustering unit is configured to cluster the point-of-stay sequence segments after dividing, so as to acquire an activity pattern of the individual. The invention further provides a mining method of activity patterns of individuals. The mining apparatus and method are suitable for mining the activity modes of the individuals, within non-fixed length, and for mining more comprehensive activity patterns of the individuals, which are representative.