BIG DATA-BASED COLUMN DATA PROCESSING METHOD, APPARATUS, AND MEDIUM

摘要

A big data-based column data processing method, an apparatus, and a medium. The big data-based column data processing method comprises: acquiring a column data set to be processed, performing, according to data attributes of column data in the column data set, classification processing on the column data, and obtaining at least two initial column data sets (110); performing unsupervised clustering processing on each of the at least two initial column data sets, and obtaining at least two unsupervised-clustered clusters, wherein the at least two unsupervised-clustered clusters have a one-to-one correspondence with the at least two initial column data sets (120); generating multiple column data pairs separately corresponding to the at least two unsupervised-clustered clusters, and determining a column name similarity level and a column comment similarity level between two pieces of column data in each of the column data pairs (130); and determining a similarity level of each of the column data pairs according to the column name similarity level and the column comment similarity level (140).