A real coded MOGA for mining classification rules with missing attribute values
Dipankar Dutta, Paramartha Dutta · 2011
Classification rule mining is one of the important data mining tasks. Optimized Rule Set (ORS) generation is a major challenge. Multi Objective Genetic Algorithm (MOGA) has been used to search available data effectively and among many objectives instead of single objective with its real coded elitist version along with special operator. Some Data Sets (DSs) having missing attribute values. In some of their earlier work researchers are either considered DS without any missing attributes values or eliminated records having missing attribute values at data preprocessing phase, considered missing values as one category of value, replaced missing values with the most common value of the attribute or assigned probability to each of the possible values to replace missing values. In this work these are not required. During training and testing phase attributes having valid values have been used for ORS generation and testing.