On the Analysis of Work Accidents Data by Using Data Preprocessing and Statistical Techniques

摘要

The volume of data used in research has increased considerably with the development of information technology. Nowadays, these data are expressed in terms of terabytes while suffering data shortage many years ago. It is necessary to overcome through the data preprocessing stage before using it in machine learning applications. The missing, noisy and inconsistent variables in the dataset are detected and the dataset are fitted by preprocessing phase. In this study, the work accident data was passed through the data preprocessing step and then univariate frequency and cross tabulation analysis were performed on these data. According to the experimental results, high risk variables have been determined in order to get the job accidents.