摘要

面對新冠疫情對全球的肆虐長達近兩年時間，反映出人類涉及活動空間與環境的擴展，接觸到的天然氣候條件更為廣泛。庇護人類機體面臨千變氣象侵襲即是穿著服飾的最主要功能，服飾舒適度會影響使用者穿著的體驗。隨著時代的推移和發展，現代人對於衣服的舒適感要求更高。很多衣服配備保護功能，強調舒適性衣服就是以此為基礎製作的趨勢。穿著服裝的舒適度並非與生俱來，決定穿著的舒適因素有三個大方向:熱生理學舒適度、人體工學舒適度以及皮膚觸感舒適度。光體積描記法（Photoplethysmography, PPG）是臨床上常見的生理訊號之一，PPG訊號結合心率變異度分析之後可以進行自律神經系統的活性評估。舒適度雖然是主觀的感受，但是生理上對於環境的變化仍是客觀的反應，因此本研究使用PPG訊號與分析作為基礎，利用機器學習進行舒適度的分類與預測。根據實驗結果顯示，機器學習模型可以提供客觀指標以輔助使用者挑選適當的服飾。

關鍵詞:服飾舒適度、光體積描記法訊號、心率變異度

Abstract

In the past two years, the COVID-19 epidemic has ravaged the world, reflecting the expansion of human activity space and environment, and the wider range of natural climate conditions that humans are exposed to. Protecting the human body from the effects of the ever-changing weather is the most important function of wearing clothing, and the comfort of the clothing will affect the user's wearing experience. With the passage of time and development, modern people have higher requirements for the comfort of clothing. Many clothes have protective functions, and the emphasis on comfortable clothes is the trend of making clothes based on them. The comfort of wearing clothing is not inherent. There are generally three directions that determine the factors of wearing comfort: thermal physiological comfort, ergonomic comfort and skin tactile comfort. Photoplethysmography (PPG) signal is one of the common physiological signals in clinical practice. The activity of the autonomic nervous system can be assessed when the PPG signal is combined with heart rate variability analysis. Although comfort is a subjective feeling, it is still an objective physiological response to environmental changes. Therefore, this study uses machine learning to classify and predict comfort based on PPG signals and analysis. According to the experimental results, the machine learning model can provide an objective indicator to help users find suitable clothing.