Apparatus and method for generating data set

1. An apparatus for generating a data set, comprising:

one or more processors; and

executable memory for storing at least one program executed by the one or more processors,

wherein the at least one program is configured to

classify collected data into numerical feature data and categorical feature data using a filter method, the collected data including network traffic data, system log data, and security event data,

perform correlation analysis on the numerical feature data and the categorical feature data using an analysis of variance (ANOVA) method and a Chi-Squared method,

generate a data set for supervised learning and a data set for unsupervised learning using correlation scores calculated through the correlation analysis, and

generate a neural network model for cyber breach threat detection based on the data set for supervised learning and a data set for unsupervised learning,

wherein the at least one program is configured to:

rank importance of features according to predefined feature criteria using the filter method and measure a correlation between data features based on the ranked importance of the features, thereby classifying the collected data into the numerical feature data and the categorical feature data,

normalize the numerical feature data using a min-max scaling method and convert the categorical feature data into numerical values using a one-hot encoding method, and

determine that data corresponds to the data set for supervised learning as a correlation score calculated using the ANOVA method is higher and a correlation score calculated using the Chi-Squared method is lower.

2. The apparatus of claim 1 , wherein the at least one program determines that data corresponds to the data set for unsupervised learning as the correlation score calculated using the ANOVA method is lower and the correlation score calculated using the Chi-Squared method is higher.

3. A method for generating a data set, performed by an apparatus for generating a data set, comprising:

classifying collected data into numerical feature data and categorical feature data using a filter method, the collected data including network traffic data, system log data, and security event data;

performing correlation analysis on the numerical feature data and the categorical feature data using an analysis of variance (ANOVA) method and a Chi-Squared method;

generating a data set for supervised learning and a data set for unsupervised learning using correlation scores calculated through the correlation analysis; and

generate a neural network model for cyber breach threat detection based on the data set for supervised learning and a data set for unsupervised learning,

wherein classifying the collected data comprises ranking importance of features according to predefined feature criteria using the filter method and measuring a correlation between data features based on the ranked importance of the features, thereby classifying the collected data into the numerical feature data and the categorical feature data,

wherein performing the correlation analysis comprises:

normalizing the numerical feature data using a min-max scaling method and converting the categorical feature data into numerical values using a one-hot encoding method; and

determining that data corresponds to the data set for supervised learning as a correlation score calculated using the ANOVA method is higher and as a correlation score calculated using the Chi-Squared method is lower.

4. The method of claim 3 , wherein performing the correlation analysis comprises determining that data corresponds to the data set for unsupervised learning as the correlation score calculated using the ANOVA method is lower and the correlation score calculated using the Chi-Squared method is higher.