Machine learning algorithm to identify eye movement metrics using raw eye tracking data

Eye-tracking studies in software engineering are becoming more prevalent and also in the areas like medical, gaming and commercial fields. Researchers may use the same metrics but it is majorly used to give a different name for same field that cause the difficulties in comparing studies, so in this work, a model is developed to reduce the existing challenges. Many existing algorithms are available to apply on eye tracking data but machine learning is one of the best algorithms, for example random forest is one the machine learning algorithms, which helps to hold the test set. In the eye movement metrics, the dataset will be divided into two sets they are: test set and training set. This paper reports on the eye-tracking metries using raw eye-tracking data. The proposed research work has used random forest, decision tree, KNN and SVM for experimentation in order to understand the dataset. The objective of this study is two-fold. First, the identification of various eye movement metrics events and Second, Apply visualization technique. It can be applied in medical field. Here first we will identify the accuracy, recall, precision and f-measure between KNN classifier and SVM, then identifying the eye movement metrics using machine learning algorithm. We give in this research a brief description of the eye movement metrics and which machine algorithm would give the best result, with its applications.