Concentration Patterns Estimation Method in Deskwork by Using Time-series k-means
Tyler Inari, Takafumi Nakanishi · 2022 International Electronics Symposium (IES) · 2022
In this paper, we present a concentration patterns estimation method in desk work using time-series k-means. Maintaining focus is very significant when working on tasks daily. Sustaining a high level of concentration helps to work more efficiently. Most people don’t recognize their concentration. Furthermore, there is no opportunity to be consciously aware of trying to recognize concentration. To improve the quality of concentration, we need to recognize concentration during work. In this paper, we propose a method to recognize and visualize concentration. Our method enables people to recognize daily concentration by clustering and visualizing concentration types. We define the word of concentration pattern is the type of user’s concentration level transition. The concentration patterns are divided into several clusters by clustering function. Each cluster has its own features, indicating that are multiple types of concentration. The visualizing function makes it easy to be aware of concentration daily. The experimental results show that our method is very effective in recognizing the concentration of the self. We also present how this method works.