This article presents an exploration of the human eye tracking data towards consumer products. The study aim to investigate the data attributes of the cognitive processes and focused on the visual attention of the participants when choosing a shampoo brand which is commonly available in Malaysia. However, eye tracking datasets has a wealth of data on the eyes visual attention, fixation, saccade and scan path gaze. Therefore, this paper aims to solve this problem to minimize the datasets by using clustering machine learning approach. This is to observe the relation of these data attributes and possibly predict the possible solution contributing to cognitive processing. Tobii TX300 Eye-tracker was used in this experiment and the eyes tracking data were gathered particularly related to the eyes fixation and saccades by using the Tobii I-VT filter. Sixty subjects participated in this study. K-means clustering was used as statistical analysis to cluster the huge datasets from the eye tracking data. The relationship of the consumer cognitive processes with visual attention was understood when most of the participants chose the most popular shampoo brand such as Head & Shoulder. Further visual analysis on the data attributes results showed that K-means clustering has the potential to cluster and minimize the huge datasets and predicts consumer preferences.