Obtain weighted values of solar stills’ environment factors by using supervised machine learning

Abstract: Solar stills have attracted increasing attentions in recent years due to its ability of simple construction and eco-friendly. In this study, the weighted values of environment factors on evaporation efficiency are obtained by using a well-established machine learning algorithm, random forest. To test the advancement between random forest and mathematical data analysis, two traditional data science methods, pair wise plots and Pearson correlation analysis, were conducted for comparison. Experimental data used in analysis were collected from around 100 articles since 2014. The results indicated that thermal design was the most significant factor that contributes in high-efficiency solar evaporation. It will promote the studies on evaporation efficiency solar stills. "