Computer-implemented method, storage medium, and computer system for parallel tree-based prediction

摘要

Embodiments relate to performing an ensemble model based prediction for a set of input data records. An aspect includes creating an assignment table assigning all data records in the input data set to a root node of a tree. Another aspect includes splitting the assignment tables into disjoint sub-tables. Another aspect includes storing each of the sub-tables to a respective data slice. Another aspect includes designating root nodes of each of the plurality of trees as a set of current nodes. Another aspect includes evaluating split conditions of each identified current node based on data records that are assigned to the current node; re-assigning said input data records to one or more child-nodes; using the child nodes of all current nodes as the set of current nodes; and calculating a prediction result from the assignment of data records to the leaf nodes of each of a plurality of trees.