DEVICE AND METHOD FOR ALGORITHM ACCURACY ENHANCEMENT BASED ON FEATURE ENGINEERING

摘要

The present invention relates to a device and a method for algorithm accuracy enhancement based on feature engineering which can provide an optimized machine learning algorithm. According to an embodiment of the present invention, the device for algorithm accuracy enhancement based on feature engineering comprises: a data receiving unit to receive data to be analyzed; a feature variable extraction unit to extract a feature variable from the data to be analyzed based on feature engineering; a calculation unit to calculate result data for an input of the feature variable from a plurality of machine learning algorithms selected in response to an analysis purpose in accordance with the feature variable; an evaluation unit to calculate evaluation indices of the plurality of machine learning algorithms based on the result data; and a visualization unit to output the feature variable and visualize the evaluation indices of the plurality of machine learning algorithms. The feature variable extraction unit can extract the feature variable from structured data or unstructured data included in the data to be analyzed, and can extract the feature variable based on a natural language processing algorithm based on unsupervised learning in the case of unstructured data.;COPYRIGHT KIPO 2020