STATISTICAL ANALYSIS FUNCTION RECOMMENDATION SYSTEM IN ACCORDANCE WITH TABLE STRUCTURE AND DATA CHARACTERISTICS

摘要

The present invention relates to a statistical analysis function recommendation system in accordance with a table structure and data characteristics which uses characteristic information for each column extracted by exploratory data analysis for an inputted table and meta data information for the table to perform learning for each analysis function, and recommends a function having a maximum score by score calculation for each statistical analysis function for the corresponding table as a statistical analysis function. According to the present invention, a statistical analysis expert matches statistical analysis functions suitable for a table, and performs learning for characteristics and a structure of the corresponding table and an analysis function to automatically recommend a statistical analysis function suitable for analysis of an inputted table by characteristic extraction for each column for the corresponding table. Also, if a client and a server system are included to continuously accumulate learning results in a server, a statistical analysis function suitable for analyzing an inputted table can be automatically recommended based on the learning results stored in the server even if the system is installed in any client device to operate.;COPYRIGHT KIPO 2017