Investigating the relationship between violations and the individual information of NAJA staff using the data mining method

Employee malfeasances are one of the most important issues of an organization human resources management. The lack of a mechanism to identify abusive employees causes the organization to face various challenges. Using data mining methods technicians can benefit from the bulk of information gathered by organizations. The purpose of this study is to investigate the relationship between violations and the individual information, using data mining method. This research is applied in terms of purpose and descriptive – analytical in terms of method, based on data mining procedure. There are 43 violations in the NAJA level. Five of these violations, which are more frequent than the incidence rate or objective statistical community, are referred to as "common violations" and include: 1) absenteeism 2) laziness and laxity in the performance of duties.  3) not to be on time at service place 4) failure to observe laws and regulations 5) lack of responsibility. The population of this research, Data on common misconduct of NAJA staff and sample size Data is 1395 years. Investigating different methods of data mining showed that the best method for the existing data and analysis of employee violations is the method of decision tree C5. In the research process, data is divided into experimental and educational categories and based on this prediction accuracy was determined in different situations. The results showed that the most prominent feature of employees in the occurrence of an offense is in the order of importance of (1) category 2) service unit and 3) service age.